NHPoisson - An R Package For Fitting and Validating Nonhomogeneous Poisson Processes

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/273888088
NHPoisson: An R Package for Fitting and Validating Nonhomogeneous

Poisson Processes
Article in Journal of Statistical Software · March 2015

DOI: 10.18637/jss.v064.i06
CITATIONS READS
5 756
3 authors, including:
Ana C. Cebrian Jesús Asin

University of Zaragoza University of Zaragoza
23 PUBLICATIONS 219 CITATIONS 45 PUBLICATIONS 345 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Statistical modelling of extreme temperature events View project
All content following this page was uploaded by Jesús Asin on 25 April 2015.
The user has requested enhancement of the downloaded file.

JSS Journal of Statistical Software
March 2015, Volume 64, Issue 6. http://www.jstatsoft.org/
NHPoisson: An R Package for Fitting and Validating

Nonhomogeneous Poisson Processes
Ana C. Cebrián Jesús Abaurrea Jesús Ası́n

University of Zaragoza University of Zaragoza University of Zaragoza
Abstract
NHPoisson is an R package for the modeling of nonhomogeneous Poisson processes in
one dimension. It includes functions for data preparation, maximum likelihood estima-
tion, covariate selection and inference based on asymptotic distributions and simulation
methods. It also provides specific methods for the estimation of Poisson processes result-
ing from a peak over threshold approach. In addition, the package supports a wide range
of model validation tools and functions for generating nonhomogenous Poisson process
trajectories. This paper is a description of the package and aims to help those interested
in modeling data using nonhomogeneous Poisson processes.
Keywords: nonhomogeneous Poisson process, peak over threshold, validation analysis, covari-
ate selection.
1. Introduction
Despite the large number of potential applications of Poisson processes, the scientific com-
munity in many fields has not yet taken full advantage of Poisson process modeling, par-
ticularly in a nonstationary framework. Most of the published applications are restricted
to the environmental field, see the review of extreme event studies on hydrology by Katz,
Parlange, and Naveau (2002). Ogata (1988) used Poisson processes to model earthquakes,
and Madsen, Rasmussen, and Rosbjerg (1997) and Abaurrea and Cebrián (2002) modeled
hydrological and meteorological drought occurrences, respectively. The statistical techniques
for modeling and fitting this type of processes are well-known, see Embrechts, Klüppelberg,
and Mikosch (1997), Coles (2001), or Kutoyants (1998) who discuss the estimation theory
for nonhomogeneous Poisson processes. However, the implementation of these techniques is
a time-consuming and non trivial task and the tools for validating nonhomogeneous Poisson
processes are not as well developed as for other statistical models. Accordingly, an easy-
2 NHPoisson: Nonhomogeneous Poisson Processes in R
to-use software, which provides simple and fast algorithms that facilitate the modeling and
validation of nonhomogeneous Poisson processes, can be very useful.
NHPoisson (Cebrian 2015) is an R (R Core Team 2015) package which provides an assembly
of tools for all the steps involved in data modeling using this type of processes. In particular,
it supplies functions to model the Poisson intensity as a function of covariates, including
the estimation of Poisson processes resulting from a peak over threshold approach. It also
contains a wide toolkit for model selection and validation analysis, including several types
of residuals and diagnostic techniques adapted from other statistical models. NHPoisson is
available from the Comprehensive R Archive Network (CRAN) at http://CRAN.R-project.
org/package=NHPoisson.
The outline of the paper is as follows. Section 2 introduces Poisson processes and the peak
over threshold approach. Some R packages related to this topic are reviewed in Section 3, and
Section 4 describes the main functions and capabilities of the package. Section 5 shows an
example of data analysis using NHPoisson: the modeling of the occurrence of extreme heat
events in a daily temperature series. Finally, some conclusions and future work are discussed
in Section 6.
2. Poisson processes and peak over threshold models

A Poisson process (PP in short) is a point process, i.e., a random collection of points in a space
where each point represents the occurrence of an event. Time processes are the most common,
but PPs can also model events in space or in space-time. The PP assumes that points occur
randomly at a given intensity λ (> 0), which characterizes the frequency these events are
expected to occur with. The model allows us to calculate values such as the probability of
observing a certain number of events in a time period, or the expected time between two
events. Examples of events that can be modeled as a PP include the occurrence of extreme
environmental events such as fires, earthquakes or heat waves, the times of customer arrivals
at queues, the occurrence times of claims in a given period, etc. Herein, we will restrict our
attention to the modeling of one-dimensional PPs, in particular, to time PPs.
Statistical modeling based on a PP requires the estimation of the intensity of the process.
A realization of a time PP is a sequence of points (ti )ni=1 observed in a region of R+ . The
information provided is that events occur at those points and nowhere else. In most of the
real problems, the intensity is time-varying and a nonhomogeneous Poisson process (NHPP)
must be considered. The behavior of a time NHPP is generally modeled by representing the
intensity λ(t) as a function of covariates, that can include time trends, seasonal terms and/or
external factors. In order to guarantee that λ is positive, a logarithm link is usually used,
log(λ(t; β)) = X> (t)β
where X> (t) is the row vector of covariates at time t and β the vector of parameters.
PPs often appear in the framework of peak over threshold (POT) models. POT models
consider the occurrence times of peaks in a sequence of random variables, where a peak is the
maximum value in a run of consecutive excesses (observations over a threshold). A well-known
result of extreme value theory states that, in an i.i.d. series, the occurrences of excesses over
a sufficiently extreme threshold u follow a PP, and the excesses have a generalized Pareto
Journal of Statistical Software 3
distribution. More precisely, given a series of i.i.d. random variables Y1 , Y2 , . . ., the process
i

Nn = , Yi , i = 1, . . . , n ,
n+1
on regions of the form (0, 1] × [u, ∞) is approximately a PP for sufficiently high u.

Although these results require i.i.d. series (Yi ), most of them are also applicable, under quite
mild conditions, to processes with short-term dependence (Coles 2001).
3. Point processes and POT in other R packages

There are several R packages dealing with point processes. Package ppstat (Hansen 2013)
focuses on the analysis of marked point processes with discrete marks. It fits generalized
linear point process models, including the Hawkes and the Gibbs families, where the inten-
sity is formulated in terms of a linear combination of linear filters. No validation tools are
implemented.
The extensive package spatstat (Baddeley and Turner 2005; Baddeley, Turner, Mateu, and
Bevan 2013) is a package for analyzing spatial point patterns. It supports the model fitting of
spatial point processes, in particular Poisson processes, and related inference and diagnostic
tools. The function ppm fits a spatial point process to an observed point pattern and allows
the inclusion of covariates. Two estimation methods, pseudo-maximum likelihood and Huang-
Ogata approximate maximum likelihood, are implemented. The available functions and data
types (point patterns, spatial windows, pixel images, etc.) are intended for spatial point
processes. Some of its functions could be applied to one-dimensional data, but the analysis of
time point processes has their own peculiarities, and the spatial framework, in particular the
specification of the points of the process, is unnecessarily complicated for the one-dimensional
case. The package yuima (Brouste et al. 2014) develops tools for inference and simulation of
stochastic differential equations to model random evolution along continuous, or practically
continuous, time. A wide variety of time processes, including compound Poisson processes,
can be fitted using a quasi-likelihood approach. Different tools for model selection are available
in the package, but the inclusion of covariates is not implemented.
There are also some R packages devoted to POT and extreme analysis, such as evir (Pfaff,
McNeil, and Stephenson 2012), POT (Ribatet 2012), and ismev (Heffernan and Stephenson
2014) and its graphical user interface extRemes (Gilleland and Katz 2011). The functions
fitpp and pot, from POT and evir respectively, fit the exceedances above a threshold, in
the homogeneous case, using the point process characterization as described in Coles (2001).
This approach requires both the occurrence times and the exceedance values. The function
pp.fit from ismev fits the point process model with a generalized linear modeling of each
parameter, by maximum likelihood. There are a wide set of tools in these packages for
selecting an adequate threshold to apply the POT model, such as the functions meplot (in
evir), mrl.plot (in ismev) or tcplot and mrlplot (in POT). Since these functions cover the
analysis of threshold selection, no function of this type is included in NHPoisson.
As far as we know, no commercial statistical software, such as SPSS (IBM Corporation 2013)
or Minitab (Minitab Inc. 2010), supports the fit of NHPPs.
4. Features and capabilities

NHPoisson aims to provide a global framework for the modeling of data using NHPPs in
one dimension. One of the new functionalities is the maximum likelihood estimation of PP
models with the intensity modeled as a function of time-dependent covariates. In particular,
it includes a specific likelihood function to estimate the PP resulting from the POT approach,
with the peculiarity that it uses information only on the point occurrences, and not on the
exceedance values. This likelihood function can also handle missing values. Another capability
is the estimation of models with fixed β parameters. The package provides automatic tools
for model selection by AIC and likelihood ratio tests, and a comprehensive toolkit for residual
analysis and diagnostics to check the fitted model. Random generation of NHPPs with a
given time-varying intensity and inference methods based on asymptotic distributions and
simulation are also implemented.
4.1. Data and exploratory analysis

The usual input to fit a time PP is the vector of points containing the occurrence times of
the events (ti )ni=1 . Since the PP is often used in the framework of POT models, the function
fitPP.fun also provides the possibility of using as input the original series (yi )Ti=1 and the
threshold which defines the extreme events. The function POTevents.fun calculates some
characteristics of these extreme events: the first and the maximum excess positions, the mean
excess, the maximum excess and the length of the events. The maximum excess position
defines the occurrence point in the point process. Missing values in the original series are not
allowed in this function, since it is difficult to adopt a general rule about how to handle them,
specially in the case of missing information between two exceedances.
The functions emplambda.fun and emplambdaD.fun calculate the empirical occurrence rates
on overlapping and disjoint intervals, respectively. The occurrence rate is calculated as the
number of points in the considered interval divided by its length, and each rate is assigned
to the mean point of the interval. Overlapping intervals are defined by a constant length
(argument lint), while disjoint intervals can be specified by the number of intervals (argument
nint) or by a constant length lint. All the intervals have the same length except the last
one, which will be shorter if the length of the series is not a multiple of lint or nint. In both
cases, a plot of the empirical rate over time can be optionally performed.
4.2. Fitting the model

The function fitPP.fun fits a NHPP by conditional maximum likelihood, given the covariates.
It is worth to recall that efficient estimation based only on the conditional model, ignoring
the information of the marginal process, can only be carried out if weak exogeneity holds
(Ericsson and Irons 1994). The conditional likelihood function is,
Z n
Y
L (β; (ti )ni=1 ) = exp − λ(t; β)dt λ(ti ; β),
A i=1
where λ(t; β) = exp(X> (t)β) and A is the space where the point process is defined. Assuming
that λ(t; β) is constant in each time unit, the expression of the loglikelihood is
T
X n
X
LL (β; (ti )ni=1 ) = − λ(t; β) + log λ(ti ; β),
t=1 i=1
where T is the length of the observation period.

Some adjustments are necessary to define the likelihood function of a NHPP based on the
POT approach. Only one occurrence point can be assigned to each event, and that is the point
of maximum intensity. The other observations in the extreme event cannot be considered as
non occurrence points, since their covariates would show their extreme character. To avoid
this problem, these observations are eliminated from the estimation sample. The result is the
likelihood we would have obtained, if only the point of maximum intensity in each event had
been observed. This is done using an index variable inddat, equal to one in the observations
to be used and zero otherwise.
This index also allows us to handle missing values by assigning them a value zero. Conse-
quently, all the functions with the argument inddat, such as fitPP.fun, emplambda.fun or
emplambdaD.fun, can deal with missing values. The function fitPP.fun can also fit homoge-
neous Poisson processes (HPP) by defining the intensity with an intercept and no covariate.
The maximum likelihood (ML) estimation can be done using two different optimization rou-
tines, optim or nlminb, selected in the argument minfun. This is useful since, in some cases,
one routine can succeed to converge when the other fails. fitPP.fun also allows us to fit a
model keeping fixed some β parameters (offset terms) using the fixed argument. This can
be used to specify an a priori known component to be included in the linear predictor.
The output of fitPP.fun is an object of the S4 class ‘mlePP’, which encapsulates the most
relevant information about the maximum likelihood estimation of a NHPP. Some of the slots
of this class are: the maximum value of the loglikelihood function, the ML estimators of the
β parameters, their covariance matrix, the fitted values λ̂(t), and the lower and upper limits
of the confidence intervals for each λ(t), whose calculation is described below. The estimation
of the β̂ covariance matrix is based on the asymptotic distribution of ML estimators and
calculated as the inverse of the negative of the Hessian,
 −1
∂ 2 LL(β)
V̂ (β̂) =  −  ,
∂βl ∂βk β=β̂
where the second derivative of the considered loglikelihood is

T
∂ 2 LL(β) X
=− λ(t; β)Xl (t)Xk (t).
∂βl ∂βk t=1
The class ‘mlePP’ is a subclass of the S4 class ‘mle’ defined in stats4, and most of the generic
functions with ‘mle’ methods, such as logLik, extractAIC or summary, can be applied to
the output of fitPP.fun. Some other generic functions related to fitted models, such as AIC
or BIC, can also be applied to ‘mlePP’ objects. There is an ‘mle’ method for vcov, which
calculates the approximate variance-covariance matrix of β̂. However, the solve function
used to obtain the inverse of the negative of the Hessian matrix may not work, specially when
its dimension is high. For this reason, we have implemented the function VARbeta.fun, which
works in a wider range of situations, since when solve gives an error, the calculation of the
inverse via its Cholesky decomposition is tried.
The package also provides a method for ‘mlePP’ objects for the generic function profile, very
similar to the ‘mle’ method for this function in stats4. As in the ‘mle’ method, the output is
a ‘profile.mle’ object, and the ‘mle’ method for confint can be used to obtain confidence
intervals based on profile likelihoods of the β parameters. Additionally, confidence intervals

based on the asymptotic normal approximation of the ML estimators of the β parameters are
calculated by the function confintAsin.fun.
Confidence intervals for the intensity

Confidence intervals for λ(t) = exp(ν(t)) can be obtained by transforming the confidence
intervals for ν(t) = X> (t)β. Another option is based on the delta method, but this approach
uses an approximation and does not always guarantee the confidence level of the interval. Both
alternatives can be calculated by fitPP.fun, which calls either the function CItran.fun or
CIdelta.fun depending on the argument CIty = "Transf" or CIty = "Delta". In general,
the transformation approach (the default) should be preferred.
The transformation approach applies an exponential transformation to the confidence interval
of the linear predictor, ν̂(t) ± z1−α/2 s.e.(ν̂(t)), where s.e.(ν̂(t)) is obtained from,
V̂ (ν̂(t)) = X> (t)V̂ (β̂)X(t).
Given that λ̂(t) is the ML estimator of λ(t) (due to the functional invariance of ML estimators),
another confidence interval of λ(t) is λ̂(t)±z1−α/2 s.e.(λ̂(t)). s.e.(λ̂(t)) is obtained by the delta
method (Casella and Berger 2002), which applied to exp(X> (t)β̂) gives
C X
X C
V̂ [λ̂(t)] = λ̂(t)Xl (t)λ̂(t)Xk (t)V̂ [β̂l , β̂k ]. (1)
l=1 k=1
That interval might include values outside the range [0, ∞), so the lower limit is defined as
the maximum between 0 and λ̂(t) − z1−α/2 s.e.(λ̂).
The function fitPP.fun optionally performs, with the argument dplot = TRUE, a plot of the
fitted intensities together with the confidence intervals for each λ(t).
Model selection
There are different tools for aiding in model selection. When the number of covariates
is high, automatic procedures such as a stepwise selection can be useful. The function
stepAICmle.fun performs a stepwise model selection by AIC for ‘mlePP’ objects. Three
directions, "forward", "backward" and "both", are implemented. The initial model is spec-
ified by the argument ImlePP, and the algorithm stops when, according to the AIC, none of
the covariates eliminated from the model or added from the potential covariates set (argument
covariatesAdd) improves the model fitted in the previous step. The definition of AIC uses
constant k = 2, but a different value k can be passed as an additional argument.
Authors such as Cook and Weisberg (1999) warn that automatic model selection methods
must be used with caution since they can reject reasonable submodels from further consid-
eration. For a more thorough selection, it can be useful to perform the likelihood ratio test
by the function testlik.fun, which compares a general model ModG, with a reduced model
ModR (a particular case of the first).
To make the selection procedure easier, the function LRTpv.fun calculates, for each covariate
in a model, the p value of a likelihood ratio test comparing the fitted NHPP with the model ex-
cluding that covariate from the linear predictor. The functions dropAIC.fun and addAIC.fun
(both used in stepAICmle.fun) perform a similar task but using the AIC value. The first
calculates the AIC for all one-covariate deletions from the current model, while the second
calculates it for all one-additions from the covariates supplied in covariatesAdd. In both
cases, the best covariate to be dropped from (added to) the model is returned together with
some additional information. All these automatic selection functions are not implemented for
models with fixed parameters.
4.3. Generation of NHPP trajectories and simulation based inference

The generation of trajectories from a NHPP is a useful tool, particularly for simulation based
inference. There exist quite a few methods to generate the points from a NHPP. One of the
most popular is the thinning approach by Lewis and Shedler (1979), an exact method which
does not require the discretization of the intensity function. However, an inversion approach
(Çinlar 1975) has been preferred here, since it is faster when the inversion can be easily
computed, as in this case. The inversion approach is based on the fact that a NHPP can be
transformed into a unit-rate HPP, using the following time scale transformation (Daley and
Vere-Jones 2003),
Z tN H
i
tH
i = λ(t)dt. (2)
0
The first step is to generate the points tH

i in a unit rate HPP via the generation of independent
exponential distances. Then, the points tN i
H in a NHPP with intensity λ(t) are the result
H
from transforming the points ti with the inverse of the function in Equation 2. Since it is
assumed that λ(t) is constant in [t, t + 1), the points tNi
H are calculated as the lowest integer
PtN H
such that i
r=0 λ(r) ≥ tH
i , where r runs in the non-negative integer numbers.
This method is supported by the function simNHP.fun, which generates the points in a period
(0, T ). The intensity in each time of this period must be provided by the argument lambda,
and the value T is determined by the length of this intensity vector.
In order to allow the exact reproduction of the results, a seed can be set in the generation
process with the argument fixed.seed. If the argument is an integer, it gives the seed and
if it is NULL (the default option), a random seed will be used.
Simulation based inference

Inference on the β parameters and on λ(t) intensities can be carried out using likelihood ratio
tests, as indicated in Section 4.2. However, there are other values whose inference is difficult
to be performed using exact or approximate distributions. In those cases, simulation based
inference is a useful too to calculate, for example, envelopes for the number of events to occur
in a given time interval, or for the time of the k-th occurrence. The upper and lower limits of
these envelopes are the m lowest and m highest values of a simulated sample. These envelopes
are not real confidence intervals for the value of the function, but significance bands that
specify the critical points for a Monte Carlo test with a significance level α = 2m/(1 + niter ),
where niter is the number of simulated trajectories (Ripley 1981).
The function GenEnv.fun calculates simulated envelopes for the value of the function specified
in the argument fun.name, which must be a function of the NHPP points and any additional
argument. It calls the function funSim.fun (not intended for the users), which generates a
simulated sample and evaluates the function in that sample. As in simNHP.fun, the argument
fixed.seed allows us to set the seed in the simulation process.
4.4. Model validation

NHPoisson includes the function globalval.fun which carries out a complete standard anal-
ysis to check the fitted NHPP. However, users can customize their own validation analysis
using the available individual functions described in this section.
Two types of residuals can be calculated: uniform (or exponential) and raw residuals. Both
of them are useful and provide complementary information.
Exponential or uniform residuals

A usual approach to validate a NHPP with intensity λ(t) is to transform it into a unit-rate
HPP, using the time transformation in Equation 2. Then, the validation analysis consists in
analyzing the inter-event distances d∗i = t∗i − t∗i−1 that, under the null hypothesis, must be
an i.i.d. exponential sample. Equivalently, the validation can be based on the transformed
distances exp(−d∗i ) that must be an i.i.d. uniform sample. These residuals may be plotted
against covariates, or potential covariates, as an informal assessment of the need of including
them. For more details on this topic, see the diagnostic analysis for Cox-Snell type residuals
in Collett (1994).
Hence, the validation analysis starts by applying the function transfH.fun to transform
the NHPP into a HPP. Then, the exponential and uniform residuals of the homogeneous
process are calculated using unifres.fun. The function graphresU.fun checks the uniform
behavior and the serial correlation in the uniform residuals. To analyze the correlation, the
function calculates the Pearson correlation coefficient, the Ljung-Box p values and a lagged
serial correlation plot. To analyze the uniform behavior, a Kolmogorov-Smirnov test and a
uniform qqplot with a 95% confidence envelope are performed. In addition, an index plot of
the residuals and residual plots against the variables specified in the argument Xvariables
are drawn.
Raw residuals
A complementary approach to validate a NHPP is the analysis of the raw residuals, see Lewis
(1972) and Baddeley, Turner, Møller, and Hazelton (2005) for this type of residuals
Rt
in time
and space-time processes. The error (or innovation) process is ε(t) = N (t) − 0 λ(u)du which,
when the model is true, must be a zero-mean martingale. The corresponding raw residual
process is
X Z l
R(l) = Iti − λ̂(u)du,
0
ti ∈(0,l)
with Iti an indicator variable, equal to 1 at time ti . The residuals are defined by the increments
of the raw process in intervals (l1 , l2 ), which can be disjoint or overlapping, divided by its
length,
 
Z l2
1 1  X
r(l1 , l2 ) = (R(l2 ) − R(l1 )) = It − λ̂(u)du . (3)
l2 − l1 l2 − l1 t ∈(l ,l ) i l1
i 1 2
These residuals can be interpreted as observed minus fitted value residuals and they are useful
to check the fitted PP intensity since, under the correct model, r(l1 , l2 ) ≈ 0.
The functions CalcRes.fun and CalcResD.fun calculate the raw residuals using overlapping
and disjoint intervals (l1 , l2 ) respectively. The residuals ro(l1 , l2 ) are based on overlapping
intervals of a given length, centered on each time t. It is noteworthy that, due to their
overlapping structure, they will not be independent even under the model. The residuals
rd(l1 , l2 ) are calculated on disjoint intervals of equal length, and assigned to the mean point
of the interval. In both functions, the empirical and the mean fitted intensities used to
calculate the raw residuals, ti ∈(l1 ,l2 ) Iti /(l2 − l1 ) and ll12 λ̂(u)du/(l2 − l1 ), are elements of the
P R
output list. Scaled versions of the residuals,

 
Z l2
1  X
rsca (l1 , l2 ) = h(ti )Iti − h(u)λ̂(u)du , (4)
l2 − l1 t ∈(l ,l ) l1
i 1 2
can be calculated using a non

q negative weight function h(u). By default, the Pearson residuals
are calculated, h(u) = 1/ λ̂(u), but any other weight function can be provided through the
argument h. The function graphrate.fun plots, for comparative purposes, the empirical and
the mean fitted intensities against time in the same graph.
A usual diagnostic tool for analyzing the covariate effects is a residual plot. In the case of
raw residuals, which are not linked to a time instant but to an interval, two different plots
can be used depending on the characteristics of the covariate: the time residual plot or the
lurking variable plot. A qqplot to check the residual distribution is also implemented.
Time residual plots The function graphres.fun plots any type of residuals (raw or scaled,
overlapping or disjoint) against time or a specific type of variables. In these plots, the residuals
are plotted against the value of the variable in the mean point of the interval and, consequently,
they are adequate only for variables which are lineal or, at least, monotonous functions of
time. A lowess smoother of the residuals can be optionally plotted to make the interpretation
easier. In the case of overlapping intervals, the residuals of the occurrence points are marked
with different symbols.
Approximate confidence envelopes can be optionally plotted in these graphs. These envelopes
are based on the approach suggested by Baddeley et al. (2005) for spatial inhomogeneous
Poisson processes. Their interpretation, including the pointwise significance, is similar to the
limits ±2 used for the standardized residuals in linear regression analysis. The envelopes are
based on the variance of the error process ε(l1 , l2 ) = ε(l2 ) − ε(l1 ). The variance of the raw
error processes for NHPPs is
Z l2
V [εr (l1 , l2 )] = λ(u)du,
l1
and for the scaled Pearson error process
V [εP (l1 , l2 )] = |l2 − l1 |.
These variances can be estimated by plugging into the expressions the fitted intensity λ̂(u).
It is noteworthy that the variance of the error V̂ [ε(l1 , l2 )] will overestimate the variance of the
q
residual process and, consequently, ±2 V̂ [ε(l1 , l2 )] will be wider than a confidence interval
based on the residual process variance. For the raw residuals defined in (3), the envelopes
are,
2
s X
± λ̂(i),
l2 − l1 i∈(l ,l )
1 2
where i runs over the integers in (l1 , l2 ). Analogously, for the Pearson residuals in Equation 4,
2
±√ .
l2 − l1
Lurking variable plots A possible way to analyze the relationship between the residuals
linked to intervals and a covariate which is not a monotonous function of time is to plot
them against the mean value of the variable in that interval. A more complicated but better
approach is to calculate the residuals on intervals defined by the values of the covariate instead
of by time. The following approach is a modification of the lurking variable plot by Baddeley
et al. (2005). The intervals considered to calculate the residuals for a covariate X(t) are
W (PX,i , PX,i+1 ) = {t : PX,i ≤ X(t) < PX,i+1 },
where PX,i is the sample i-percentile of X. The function graphResX.fun plots the residuals
against the mean value of the covariate in that interval. In the case of raw or Pearson
residuals, it also calculates approximate confidence envelopes as the ones described for time
residual plots. The function graphResCov.fun performs the lurking variable plot for each
variable in the argument Xvar.
Residual qqplot A common plot to check the distributional assumptions in a model is the
qqplot. The theoretical distribution of the NHPP raw residuals is unknown but a simulation
based qqplot can be calculated. The function resQQplot.fun compares the empirical quan-
tiles of the residuals with the expected quantiles under the fitted model, obtained from the
following simulation process.
First, the model is fitted to the data set and the residuals ri are calculated. The order statis-
tics r(1) ≤ r(2) ≤ . . . ≤ r(n) are obtained from the residual sample. Then, niter independent
trajectories of a NHPP(λ̂(t)) are generated and a NHPP is fitted to each one. The correspond-
ing fitted intensities and the ordered residuals from the i-th trajectory are denoted λ̂(i) (t) and
(i) (i) (i)
r(1) ≤ r(2) ≤ . . . ≤ r(n) , respectively. The expected j-th quantile of the residuals under the
fitted model is calculated as the sample mean of the j-th order statistic
1 niter
X (i)
ej = r .
niter i=1 (j)
Individual (1 − α)% envelopes for each residual are defined by the corresponding α/2 and
1 − α/2 sample quantiles.
The function resQQplot.fun calculates for each residual the simulated expected quantile and
plots them together with the 1 − α individual envelopes. Disjoint or overlapping and Pearson
or raw residuals are supported by this function, but the overlapping option can require a high
time computing cost. The output list of the function includes the expected quantiles and the
upper and lower limits of the envelopes of the residuals. The argument fixed.seed allows us
to set the seed to obtain reproducible results.
A global validation function

A standard validation analysis to check a NHPP model can be performed in one step using
the function globalval.fun. First, the uniform residuals are calculated and analyzed. The
residuals are plotted against the covariates in Xvar and, if this argument is NULL, all the
covariates in the fitted model are considered. The empirical and the cumulative mean fitted
intensities (using optionally overlapping or disjoint intervals) are plotted together. Using
these intensities, the raw or scaled residuals are calculated, and residual plots against time
and the variables in Xvart are carried out. The lurking variable plots for the variables in
Xvar are also performed; as before, if this argument is NULL, all the covariates in the fitted
model are considered. Finally, a residual qqplot to check the distributional assumptions can
be optionally calculated. By default, the qqplot is carried out for the Pearson residuals in
disjoint intervals.
The argument histWgraph = TRUE (the default option) opens a new graphical device (using
dev.new(record = TRUE), which may not work on all systems), so that the history of all
the plots carried out by the function is recorded in the same device. This means that the
resulting plots can be accessed by scrolling up and down. The argument plotDisp provides
the layout to display the residual plots, 2 × 2 by default. These both arguments are available
in all the functions which carry out several residual plots: graphres.fun, graphresU.fun,
graphresCov.fun, etc.
5. An example
In this section, we illustrate how the package NHPoisson can be used to model a data set,
from the data preparation for a POT approach to the validation analysis. Data are daily
maximum temperature series, and our aim is to model the occurrence of its extreme heat
events using a NHPP.
5.1. Data
The data set BarTxTn is available in the package NHPoisson. The variables Tx and Tn are
the daily maximum and minimum temperature series, in tenths of ◦ C, corresponding to the
summer months May, June, July, August and September (MJJAS) from 1951 to 2004 in
Barcelona (Spain). The date (day, month and year) of the observations, and some variables
representing the short and long-term temperature evolution are also available. The daily
temperature series were provided by the Spanish Meteorological Office (AEMET).
The variables are recorded at a discrete time scale, as in most real data applications. However,
given that the time unit is short compared with the length of the recorded period and the
occurrence rate very low, the use of a continuous time model, such as the PP, is justified.
5.2. Data preparation

To define an extreme heat event (EHE) we use the POT approach. The extreme threshold
is 31.8 ◦ C, which is the 95th percentile of Tx during a reference period (months JJA in
[1971, 2000]). The selection of this threshold (not shown here) is based on climate basis
and on statistical criteria about the Poisson behavior of the resulting peak series (using for
example the functions diplot, mrlplot and tcplot of the POT package). The EHEs and
their characteristics are obtained by,
R> data("BarTxTn", package = "NHPoisson")

R> dateB <- cbind(BarTxTn$ano, BarTxTn$mes, BarTxTn$diames)
R> BarEv <- POTevents.fun(T = BarTxTn$Tx, thres = 318, date = dateB)
Number of events: 137

Number of excesses over threshold 318 : 253
where the matrix dateB contains the date of each observation, and the object BarEv several
characteristics of the 137 EHEs, such as Px, the points of maximum intensity in the extreme
events, which are the occurrence points of the process. Another element of the object is the
index inddat, which marks the observations to be used in the estimation process,
R> Px <- BarEv$Px

R> inddat <- BarEv$inddat
To describe the time evolution of the EHE occurrence, a plot of its empirical rate calculated
in 153-day long overlapping periods is performed using the following commands (153 is the
length of the period MJJAS),
R> tB <- BarTxTn$ano + rep(c(0:152) / 153,55)

R> emplambdaB <- emplambda.fun(posE = Px, inddat = inddat, t = tB,
+ lint = 153, tit = "Barcelona")
where tB is an index time of the summer months linked as consecutive time periods. The
argument inddat must be specified only when there are missing values, or when the point
process is defined using the POT approach and the length of the resulting events can be
greater than one. By default, all the inddat values are equal to 1 so that all the available
observations are included in the estimation process.
5.3. Fitting the model

Due to the characteristics of the EHE occurrence (seasonal behavior and a nonmonotonic
trend), it must be modeled by a PP with a nonhomogeneous intensity (Abaurrea, Ası́n, Ce-
brián, and Centelles 2007). This intensity is modeled as a deterministic function of covariates
such as long and short-term temperature, and seasonal terms (the part corresponding to
the summer months of the harmonic functions describing the annual cycle). Temperature-
harmonic interaction terms are also considered as possible covariates.
As an exploratory step to check which covariates are more influential, we carry out an au-
tomatic stepwise selection by AIC in both directions. First, the initial model which only
includes the intercept is fitted, and the first order harmonic terms and Txm31 , Tnm31 , TTx
and TTn temperature covariates (all available from BarT xT n) are considered as potential
covariates. In order to make the interpretation of the output easier, names are assigned to
each covariate, using dimnames. It is recalled that with the arguments modSim = TRUE, dplot
= FALSE and modCI = FALSE, the information on the screen is eliminated, the fitted intensity
is not plotted and the confidence intervals are not calculated.
R> mod1Bind <- fitPP.fun(covariates = NULL, posE = BarEv$Px,

+ inddat = BarEv$inddat, tit = "BAR Intercept", start = list(b0 = 1))
R> covB <- cbind(cos(2 * pi * BarTxTn$dia / 365),
+ sin(2 * pi * BarTxTn$dia / 365), BarTxTn$Txm31, BarTxTn$Tnm31,
+ BarTxTn$TTx, BarTxTn$TTn)
R> dimnames(covB) <- list(NULL, c("Cos", "Sin", "Txm31", "Tnm31",
+ "TTx", "TTn"))
R> aux <- stepAICmle.fun(ImlePP = mod1Bind, covariatesAdd = covB,
+ startAdd = c(1, -1, 0, 0, 0, 0), direction = "both")
Step Forward 1
Initial model. AIC: 1400.471
Initial model adding covariates
AIC
Cos 1353.504
Sin 1388.329
Txm31 1072.380
Tnm31 1125.856
TTx 1367.911
TTn 1382.534
The best covariate to add is Txm31
This covariate improves the model
Step Backward 1
Initial model 1072.38
Initial model deleting covariate
AIC
Txm31 1400.471
The best covariate to drop is Txm31
This drop does not improve the model
Step Forward 2
Initial model. AIC: 1072.38
Initial model adding covariates
AIC
Cos 1073.688
Sin 1073.257
Tnm31 1074.372
TTx 1074.379
TTn 1073.412
The best covariate to add is Sin
This covariate does not improve the model
Step Backward 2
Initial model 1072.38
Initial model deleting covariate
AIC
Txm31 1400.471
The best covariate to drop is Txm31
This drop does not improve the model
Final model
The covariates added and dropped to the initial model are:
added
"Txm31"
Summary of the model:

Maximum likelihood estimation
Call:
fitPP.fun(covariates = ..1, start = ..2, posE = BarEv$Px,
inddat = BarEv$inddat, modCI = FALSE, tit = "BAR Intercept",
modSim = TRUE, dplot = FALSE)
Coefficients:
Estimate
b0 -21.7177999
b1 0.0653722
-2 log L: 1068.38
According to the AIC, the best model should only include the short-term temperature co-
variate Txm31 , but a more thorough covariate selection is advisable. We suggest a selection
based on the likelihood ratio test, using a forward stepwise approach controlled by the user.
We start by fitting a HPP,
R> modB.1 <- fitPP.fun(covariates = NULL, posE = Px, inddat = inddat,

+ tit = "BARCELONA Tx; Intercept", start = list(b0 = 1), dplot = FALSE,
+ modCI = FALSE)
Number of observations not used in the estimation process: 116

Total number of time observations: 8415
Convergence code: 0
Convergence attained
Loglikelihood: -699.236
Estimated coefficients:
b0
-4.104
Full coefficients:
b0
-4.104
attr(,"TypeCoeff")
[1] "Fixed: No fixed parameters"
The following commands analyze if the first order harmonic terms must be included,

+ sin(2 * pi * BarTxTn$dia / 365))
R> modB.2 <- fitPP.fun(covariates = covB, posE = Px, inddat = inddat,
+ tit = "BARCELONA Tx; Cos, Sin", start = list(b0 = -100, b1 = 1,
+ b2 = 1), modSim = TRUE, dplot = FALSE, modCI = FALSE)
R> aux <- testlik.fun(ModG = modB.2, ModR = modB.1)
General Model (hypothesis H1): BARCELONA Tx; Cos, Sin

Reduced Model (hypothesis H0): BARCELONA Tx; Intercept
ML ratio test statistic: 154.19
P-value: 0
Since the p value is 0, the first order harmonic is included and the inclusion of the second
order harmonic is checked.

+ sin(2 * pi * BarTxTn$dia / 365),
+ cos(4 * pi * BarTxTn$dia / 365), sin(4 * pi * BarTxTn$dia / 365))
R> modB.3 <- fitPP.fun(covariates = covB, posE = Px, inddat = inddat,
+ tit = "BARCELONA Tx; Cos, Sin, Cos2, Sin2", start = list(b0 = -100,
+ b1 = 1, b2 = 1, b3 = 1, b4 = 1), modSim = TRUE, dplot = FALSE,
+ modCI = FALSE)
R> aux <- testlik.fun(ModG = modB.3, ModR = modB.2)
General Model (hypothesis H1): BARCELONA Tx; Cos, Sin, Cos2, Sin2
Reduced Model (hypothesis H0): BARCELONA Tx; Cos, Sin
ML ratio test statistic: 0.96
P-value: 0.619
The p value 0.619 rejects the inclusion of the second order harmonic. Analogously, long and
short term temperature signals TTx , TTn, Txm31 and Tnm31 and interaction terms between
the first order harmonic and the significant temperature covariates are analyzed. The final
model resulting from this forward covariate selection process includes the first order harmonic
term and Txm31 . To fit this model, the default values of modSim, modCI, CItype and dplot
are used and, consequently, model information is shown, transformed confidence intervals are
calculated and the fitted intensity is plotted, see Figure 1.
R> covB.final <- cbind(cos(2 * pi * BarTxTn$dia / 365),

+ sin(2 * pi * BarTxTn$dia / 365), BarTxTn$Txm31)
R> dimnames(covB.final) <- list(NULL, c("Cos", "Sin", "Txm31"))
R> modB.final <- fitPP.fun(covariates = covB.final, posE = Px,
+ inddat = inddat, tim = tB, tit = "BARCELONA Tx; Cos, Sin, Txm31",
+ start = list(b0 = -100, b1 = 1, b2 = 1, b3 = 0))
BARCELONA Tx; cos, sin, Txm31
Fitted intensity
Upper CI Transf
Lower CI Transf
1.5
1.0
intensity
0.5
0.0
1950 1960 1970 1980 1990 2000
time
Figure 1: Fitted intensity and transformed confidence intervals (Section 4.2) of the selected
NHPP model.
Number of observations not used in the estimation process: 116

Total number of time observations: 8415
Convergence code: 0
Convergence attained
Loglikelihood: -533.625
Estimated coefficients:
b0 b1 b2 b3
-21.681 0.080 0.320 0.066
Full coefficients:
b0 b1 b2 b3
-21.681 0.080 0.320 0.066
attr(,"TypeCoeff")
[1] "Fixed: No fixed parameters"
LRTpv.fun shows that the first order harmonic terms are not individually significant when
Txm31 is included in the model. However, they are kept in order to make easier the compar-
ison of the temperature effect with models for other locations.
R> aux <- LRTpv.fun(modB.final)

The p-values of the LRT comparing the initial model and the model
without the covariate
p-values
Cos 0.935
Sin 0.508
Txm31 0.000
The standard error of the estimated coefficients and a summary of the final model is obtained
using,
R> summary(modB.final)
Maximum likelihood estimation
Call:
fitPP.fun(covariates = covB.final, start = list(b0 = -100, b1 = 1,
b2 = 1, b3 = 0), posE = Px, inddat = inddat, tim = tB,
tit = "BARCELONA Tx; Cos, Sin, Txm31")
Coefficients:
Estimate Std. Error
b0 -21.68078342 1.027972827
b1 0.07971801 0.908504548
b2 0.32020449 0.447603298
b3 0.06591333 0.003559439
-2 log L: 1067.25
The 95% confidence intervals of the β parameters, based on the profile likelihood and on the
properties of the ML estimators, can be obtained with the following commands,
R> confint(modB.final)
Profiling...
2.5 % 97.5 %
b0 -24.03266857 -19.46601384
b1 -1.92800787 1.90282196
b2 -0.66536902 1.20239256
b3 0.05685904 0.07475327
R> confintAsin.fun(modB.final)
2.5 % 97.5 %
b0 -23.69557314 -19.6659937
b1 -1.70091819 1.8603542
b2 -0.55708186 1.1974908
b3 0.05893695 0.0728897
1.0 Model: BARCELONA; cos, sin, Txm31
0.2 0.4 0.6 0.8 1.0

● ●
● ● ● ●
● ● ●
● ●
● ● ●
●
● ● ●
●
0.8
● ●
● ● ●
● ●
●
● ●
● ● ● ●
● ●
● ●
residuals (t)
●
●
0.6
●
● ●
● ●
● ●●
●●
ACF
● ●
● ●●
● ●
● ●●
● ● ● ●
● ●
● ● ●●
● ●
0.4
● ● ● ● ●
● ●
● ● ●
● ●
● ●
● ● ● ● ●
● ● ●
● ● ● ● ●
●
● ● ● ●
● ●
●
0.2
● ● ●
●
● ●● ●
● ● ●
● ●
● ●
●
● ● ●
● ● ● ●
●
0.0
−0.2
●
●
0.0 0.2 0.4 0.6 0.8 1.0 0 5 10 15 20
residuals(t−1) lag
Pearson coef.: 0.02. P−value: 0.833
1.0
1.0
●●
●
●●
●●
●●
●
●
●
●●
●
●
●●
●
0.8
0.8
● ●
●●
●
●●
●
●●
●●
●●
●●
●●
●●
●●
●
0.6
p−values
0.6
●
●●
●
●●
●
unires
●●
●●
●
●●
●
●●
●
●
●●
●●
●
●
●
●●
●●
●●
●
●
0.4
0.4
●
●●
●●
●
●
●●
●
●
●
●
●●
●●
●●
●
●
● ● ● ●
●●
● ●
●
●●
●●
●●
●
● ● ● ●
●
●
●●
●
0.2
0.2
● ● ● ● ●●
●●
● ●●
●●
●
● ●●
●●
● ●
●
●●
●●
●
●
●
●●
●●
●
0.0
0.0
●
●
5 10 15 20 0.0 0.2 0.4 0.6 0.8 1.0
lag unif quantiles

Ljung−Box p−values KS (uniform) p−value: 0.359
Figure 2: Validation plots of the uniform residuals of the final NHPP model.
5.4. Model validation

The validation starts by the analysis of the uniform residuals using the commands,
R> posEHB <- transfH.fun(modB.final)$posEH

R> resB <- unifres.fun(posEHB)
R> graphresU.fun(unires = resB$unires, posE = modB.final@posE,
+ Xvariables = cbind(covB.final, BarTxTn$dia),
+ namXv = c("cos", "sin", "Txm31", "summer day index"),
+ tit = "BARCELONA; cos, sin, Txm31", addlow = FALSE)
The results of the autocorrelation analysis and the uniform behaviour are satisfactory, see
Figure 2. The plots of the uniform residuals against the covariates included in the model and
the index time corresponding to the summer cycle, not shown here, are also requested for
Model: BARCELONA; cos, sin, Txm31
Empirical rate
Fitted rate
0.15
empirical and fitted occurrence rates
0.10
0.05
0.00
1950 1960 1970 1980 1990 2000
time
Rates calculated in Disjoint intervals of length 153
Figure 3: Empirical and fitted intensities calculated on disjoint summer intervals.
analysing the linear predictor.

In order to analyse the raw residual, the scaled Pearson residuals on 153-day long disjoint
intervals are calculated as follows,
R> ResDB <- CalcResD.fun(mlePP = modB.final, lint = 153)
Number of intervals to calculate the disjoint residuals: 55 of length: 153
The output of the function CalcResD.fun is a list and one of its elements are the residuals,
which can be used in other commands, for example to analyze their normal behavior
R> qqnorm(ResDB$RawRes)
Other elements in the output list are the empirical and the fitted intensities used to calculate
the raw residuals. They can be graphically compared, see Figure 3, using the following
command,
R> graphrate.fun(ResDB, tit = "BARCELONA; cos, sin, Txm31")
The lurking variable plots of nonscaled residuals against Txm31 (a covariate included in
the model) and TTx (not included), and their confidence envelopes, are obtained from the
following code snippet,
R> covBtemp <- cbind(BarTxTn$Txm31, BarTxTn$TTx)

R> aux <- graphResCov.fun(mlePP = modB.final, nint = 50,
0.06
0.04
0.05
0.02
●
●●
● ●
● ●
Raw residuals
Raw residuals
● ●
● ● ●
●
● ● ●
● ●● ● ● ●
0.00
● ●
● ●
0.00
● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ●●● ● ● ● ● ● ● ●
● ●● ●● ● ● ● ●● ● ●● ●
●● ● ● ●
●●● ● ●
● ● ●●● ●
● ● ● ●
●
−0.02
● ●
−0.05
−0.04
−0.06
●
180 200 220 240 260 280 300 240 245 250 255 260 265
Txm31 TTx
Figure 4: Lurking plots based on raw residuals for two temperature variables.
+ X = covBtemp, namX = c("Txm31", "TTx"),

+ tit = "BARCELONA; cos, sin, Txm31", typeRes = "Raw")
The lurking plots in Figure 4 show that the influence of Txm31 is properly captured by the
model and that the inclusion of TTx is not necessary. If we want to obtain plots displayed
individually (but recorded on the same graphical device), the arguments plotDisp=c(1,1)
and the default option histWgraph=TRUE should be used.
The time residual plot for the ResDB residuals and the residual qqplot are carried out using
the following commands (in resQQplot.fun the seed is set to obtain reproducible results.),
R> graphres.fun(objres = ResDB, typeRes = "Pearson", addlow = TRUE,
+ plotDisp = c(1, 1), tit = "BARCELONA; cos, sin, Txm31")
R> aux <- resQQplot.fun(nsim = 100, objres = ResDB, covariates = covB.final,
+ tit = "BARCELONA; cos, sin, Txm31", fixed.seed = 123)
The plots do not show any evidence against the fitted model, see Figure 5. It is noteworthy
to recall that depending on the number of points in the process, the number of simulations
and the computer specifications, the computation time of the function resQQplot.fun can
be quite high. In any case, at least 100 simulations are recommended.
All the previous analysis can be obtained in one step using the globalval.fun,
R> aux <- globalval.fun(mlePP = modB.final, lint = 153,
+ typeI = "Disjoint", typeResLV = "Raw", nintLP = 50,
+ tit = "BARCELONA; cos, sin, Txm31", Xvar = covBtemp,
+ namXvar = c("Txm31", "TTx"), resqqplot = TRUE, fixed.seed = 123)
Number of intervals to calculate the disjoint residuals: 55 of length: 153
5.5. Simulation based inference

Once the model is satisfactorily checked, inference can be performed. The following command
Model: BARCELONA; cos, sin, Txm31

Disjoint Pearson residuals
0.20
●
● ●
0.15
● ●
0.1
empirical quantile of the residuals

●
● ● ●
0.10
● ●
Pearson residuals
● ●
● ●
●
● ●
● ● ● ●
0.05
● ●
0.0
● ●●
● ●
● ● ● ●
● ● ●
● ● ● ●
● ● ●
● ● ● ● ● ●
●
●
●
● ●● ● ●
0.00
● ● ● ●
● ● ● ● ●
● ●
● ● ● ●
● ● ● ● ● ●
● ● ● ●
● ● ● ●
−0.1
●
● ● ●
● ●
−0.05
●
● ●
● ● ● ●
● ● ●
●
●
● ●
●
● ●
−0.10
●
1950 1960 1970 1980 1990 2000 −0.01 0.00 0.01 0.02
time mean quantile of simulated residuals
Figure 5: Time residual plot and residual qqplot (with seed 123 and calculated using one
core) for the Pearson disjoint residuals.
calculates a point estimator (the mean) and a prediction envelope for the number of occur-
rences in a summer with a given intensity function; the results are obtained with R 3.1.3 for
Windows (64 bit) and a seed equal to 123,
R> aux <- GenEnv.fun(lambda = modB.final@lambdafit[8263:8415],

+ fun.name = "length", nsim = 1000, fixed.seed = 123)
Lower interval: 1
Mean value: 3.975
Upper interval: 8
Number of valid simulations: 1000
Using the same approach, we can estimate the occurrence time of the first EHE,
R> aux <- GenEnv.fun(lambda = modB.final@lambdafit[8263:8415],

+ fun.name = "posk.fun", fun.args = 1, nsim = 1000, fixed.seed = 123)
Lower interval: 21
Mean value: 54.6383
Upper interval: 85.7
Number of valid simulations: 987
where posk.fun(x, k) returns the element in the k-th row of the vector x. The estimated
occurrence time of the first EHE in a summer with the considered intensity is the 24th of
June and the prediction envelope moves from the 21st of May to the 25th of July. It is
noteworthy that this function is estimating the occurrence time of the first EHE given that,
at least, one EHE occurs in that period. This is due to the fact that simulated trajectories
where no EHE occurs are not included in the calculations. Denoting the number of points in
the specified
R 8415
time interval by N (., .), the probability that this occurs is P (N (8263, 8415) =
0) = 8263 λ(t)dt = 0.023, practically negligible. This probability could also be estimated by
simulation using the function simNHP.fun. The simplest, but not fastest, way is
R> L <- NULL

R> for (i in c(1:1000))
+ L[i] <- length(simNHP.fun(lambda =
+ modB.final@lambdafit[8263:8415], fixed.seed = (123 + i))$posNH)
R> P0 <- sum(L == 0) / 1000
R> P0
[1] 0.014
6. Conclusions and future work

NHPoisson is an R package designed to facilitate the modeling of data using one-dimensional
NHPPs whose intensity is described as a function of covariates. This is an important feature
since the use of nonhomogeneous processes allows us to include information on the state of
the system, the geographical position of an event, or different environmental characteristics.
The package provides an assembly of all the steps involved in the modeling process: data
preparation, maximum likelihood estimation (including fixed parameters), covariate selec-
tion, inference, simulation, and a wide toolkit for validation analysis, including residuals and
diagnostic techniques adapted from other statistical models. Since the output of the main
function fitPP.fun is an S4 class object which extends the ‘mle’ class, many available meth-
ods for this type of objects and some generic functions related to fitted models can also be
used.
This package does not include techniques on threshold selection for POT models since they
are already available in other packages. However, an extension of the package could include
the implementation of new techniques on this topic, such as the ones developed by Wadsworth
and Tawn (2012).
Some future additions for the package are functions for the statistical modeling based on
multivariate nonhomogeneous Poisson processes. An example of this type of model is the
common Poisson shock process, a multivariate process with dependent components.
Acknowledgments
First, we thank all the people supporting the R project which provides a nice, powerful and
open-source environment for developing and spreading new statistical techniques, specially
the CRAN team for their efficient work. We also thank the anonymous reviewers for their
helpful comments to improve the package and the manuscript. Forthcoming suggestions from
all users are welcomed. This work was partially supported by Ministerio de Educación y
Ciencia (Spanish Department of Science) through the project CGL2009-09646.
References
Abaurrea J, Ası́n J, Cebrián AC, Centelles A (2007). “Modeling and Forecasting Extreme
Heat Events in the Central Ebro Valley, a Continental-Mediterranean Area.” Global and
Planetary Change, 57(1–2), 43–58.
Abaurrea J, Cebrián AC (2002). “Drought Analysis Based on a Cluster Poisson Model:

Distribution of the Most Severe Drought.” Climate Research, 22(4), 227–235.
Baddeley A, Turner R, Mateu J, Bevan A (2013). “Hybrids of Gibbs Point Process Models
and Their Implementation.” Journal of Statistical Software, 55(11), 1–43. URL http:
//www.jstatsoft.org/v55/i11/.
Baddeley AJ, Turner R (2005). “spatstat: An R Package for Analyzing Spatial Point Patterns.”
Journal of Statistical Software, 12(6), 1–42. URL http://www.jstatsoft.org/v12/i06/.
Baddeley AJ, Turner R, Møller J, Hazelton J (2005). “Residual Analysis for Spatial Point
Processes.” Journal of the Royal Statistical Society B, 67(5), 617–666.
Brouste A, Fukasawa M, Hino H, Iacus S, Kamatani K, Koike Y, Masuda H, Nomura R, Ogi-

hara T, Shimuzu Y, Uchida M, Yoshida N (2014). “The YUIMA Project: A Computational
Framework for Simulation and Inference of Stochastic Differential Equations.” Journal of
Statistical Software, 57(4), 1–51. URL http://www.jstatsoft.org/v57/i04/.
Casella G, Berger RL (2002). Statistical Inference. Brooks/Cole.
Çinlar E (1975). Introduction to Stochastic Processes. Prentice-Hall.
Cebrian AC (2015). NHPoisson: Modelling and Validation of Non Homogeneous Poisson

Processes. R package version 3.1, URL http://CRAN.R-project.org/package=NHPoisson.
Coles S (2001). An Introduction to Statistical Modeling of Extreme Values. Springer-Verlag.
Collett D (1994). Modelling Survival Data in Medical Research. Chapman and Hall, London.
Cook RD, Weisberg S (1999). Applied Regression Including Computing and Graphics. John
Wiley & Sons.
Daley D, Vere-Jones D (2003). An Introduction to the Theory of Point Processes. Vol I:

Elementary Theory and Methods. Springer-Verlag.
Embrechts P, Klüppelberg C, Mikosch T (1997). Modelling Extremal Events. Springer-Verlag.
Ericsson NR, Irons JS (1994). Testing Exogeneity. Oxford University Press.
Gilleland E, Katz RW (2011). “New Software to Analyze How Extremes Change over Time.”
Eos, Transactions American Geophysical Union, 92(2), 13–14.
Hansen NR (2013). ppstat: Point Process Statistics. R package version 0.9, URL http:
//CRAN.R-project.org/package=ppstat.
Heffernan JE, Stephenson AG (2014). ismev: An Introduction to Statistical Modeling of

Extreme Values. R package version 1.40, URL http://CRAN.R-project.org/package=
ismev.
IBM Corporation (2013). IBM SPSS Statistics 22. IBM Corporation, Armonk. URL http:
//www.ibm.com/software/analytics/spss/.
Katz RW, Parlange MB, Naveau P (2002). “Statistics of Extremes in Hydrology.” Advances
in Water Resources, 25(8–12), 1287–1304.
Kutoyants YA (1998). Statistical Inference for Spatial Poison Processes. Springer-Verlag.
Lewis PAW (1972). “Recent Results in the Statistical Analysis of Univariate Point Processes.”
In PAW Lewis (ed.), Stochastic Point Processes, pp. 1–54. John Wiley & Sons.
Lewis PAW, Shedler GS (1979). “Simulation of Nonhomogenous Poisson Processes by Thin-

ning.” Naval Research Logistics Quarterly, 26(3), 403–413.
Madsen H, Rasmussen PF, Rosbjerg D (1997). “Comparison of Annual Maximum Series

and Partial Duration Series Methods for Modelling Extreme Hydrologic Events. 1. At Site
Modelling.” Water Resources Research, 33(4), 747–757.
Minitab Inc (2010). Minitab 17: Statistical Software for Process Control, Improvement and
Education. Minitab Inc., State College. URL http://www.minitab.com/.
Ogata Y (1988). “Statistical Models for Earthquake Occurrences and Residual Analysis for
Point Processes.” Journal of the American Statistical Association, 83(401), 9–27.
Pfaff B, McNeil A, Stephenson AG (2012). evir: Extreme Values in R. R package version 1.7-
3, URL http://CRAN.R-project.org/package=evir.
R Core Team (2015). R: A Language and Environment for Statistical Computing. R Founda-
tion for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.
Ribatet M (2012). POT: Generalized Pareto Distribution and Peaks over Threshold. R pack-
age version 1.1-3, URL http://CRAN.R-project.org/package=POT.
Ripley B (1981). Spatial Statistics. John Wiley & Sons.
Wadsworth JL, Tawn JA (2012). “Likelihood-Based Procedures for Threshold Diagnostics

and Uncertainty in Extreme Value Modelling.” Journal of the Royal Statistical Society B,
74(3), 543–567.
Affiliation:
Ana C. Cebrián
Departamento de Métodos Estadı́sticos
University of Zaragoza
Spain
Telephone: +34/976762885
E-mail: [email protected]
Journal of Statistical Software http://www.jstatsoft.org/

published by the American Statistical Association http://www.amstat.org/
Volume 64, Issue 6 Submitted: 2013-05-07
March 2015 Accepted: 2014-07-15
View publication stats

NHPoisson - An R Package For Fitting and Validating Nonhomogeneous Poisson Processes

Uploaded by

Copyright:

Available Formats

NHPoisson - An R Package For Fitting and Validating Nonhomogeneous Poisson Processes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NHPoisson - An R Package For Fitting and Validating Nonhomogeneous Poisson Processes

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

NHPoisson: An R Package for Fitting and Validating Nonhomogeneous

Article in Journal of Statistical Software · March 2015

Ana C. Cebrian Jesús Asin

SEE PROFILE SEE PROFILE

Statistical modelling of extreme temperature events View project

The user has requested enhancement of the downloaded file.

NHPoisson: An R Package for Fitting and Validating

Ana C. Cebrián Jesús Abaurrea Jesús Ası́n

2. Poisson processes and peak over threshold models

log(λ(t; β)) = X> (t)β

on regions of the form (0, 1] × [u, ∞) is approximately a PP for sufficiently high u.

3. Point processes and POT in other R packages

4. Features and capabilities

4.1. Data and exploratory analysis

4.2. Fitting the model

where T is the length of the observation period.

where the second derivative of the considered loglikelihood is

intervals based on profile likelihoods of the β parameters. Additionally, confidence intervals

Confidence intervals for the intensity

V̂ (ν̂(t)) = X> (t)V̂ (β̂)X(t).

4.3. Generation of NHPP trajectories and simulation based inference

The first step is to generate the points tH

Simulation based inference

4.4. Model validation

Exponential or uniform residuals

output list. Scaled versions of the residuals,

can be calculated using a non

and for the scaled Pearson error process

V [εP (l1 , l2 )] = |l2 − l1 |.

W (PX,i , PX,i+1 ) = {t : PX,i ≤ X(t) < PX,i+1 },

A global validation function

5.2. Data preparation

R> data("BarTxTn", package = "NHPoisson")

Number of events: 137

R> Px <- BarEv$Px

R> tB <- BarTxTn$ano + rep(c(0:152) / 153,55)

5.3. Fitting the model

R> mod1Bind <- fitPP.fun(covariates = NULL, posE = BarEv$Px,

Summary of the model:

R> modB.1 <- fitPP.fun(covariates = NULL, posE = Px, inddat = inddat,

Number of observations not used in the estimation process: 116

R> covB <- cbind(cos(2 * pi * BarTxTn$dia / 365),

General Model (hypothesis H1): BARCELONA Tx; Cos, Sin

R> covB <- cbind(cos(2 * pi * BarTxTn$dia / 365),

R> covB.final <- cbind(cos(2 * pi * BarTxTn$dia / 365),

BARCELONA Tx; cos, sin, Txm31

1950 1960 1970 1980 1990 2000

Number of observations not used in the estimation process: 116

R> aux <- LRTpv.fun(modB.final)

Maximum likelihood estimation

1.0 Model: BARCELONA; cos, sin, Txm31

0.2 0.4 0.6 0.8 1.0

0.0 0.2 0.4 0.6 0.8 1.0 0 5 10 15 20

5 10 15 20 0.0 0.2 0.4 0.6 0.8 1.0

lag unif quantiles

5.4. Model validation

R> posEHB <- transfH.fun(modB.final)$posEH

Model: BARCELONA; cos, sin, Txm31