1 1.Definitions & examples 2.Conditional intensity & Papangelou intensity 3.Models a) Renewal processes b) Poisson processes c) Cluster models d) Inhibition.

Slides:



Advertisements
Similar presentations
Random Processes Introduction (2)
Advertisements

Spatial point patterns and Geostatistics an introduction
Spatial point patterns and Geostatistics an introduction
Exponential Distribution
Let X 1, X 2,..., X n be a set of independent random variables having a common distribution, and let E[ X i ] = . then, with probability 1 Strong law.
MOMENT GENERATING FUNCTION AND STATISTICAL DISTRIBUTIONS
Exponential Distribution. = mean interval between consequent events = rate = mean number of counts in the unit interval > 0 X = distance between events.
Statistics review of basic probability and statistics.
Chapter 5 Statistical Models in Simulation
Hydrologic Statistics
Model Fitting Jean-Yves Le Boudec 0. Contents 1 Virus Infection Data We would like to capture the growth of infected hosts (explanatory model) An exponential.
Engineering Probability and Statistics - SE-205 -Chap 4 By S. O. Duffuaa.
Simulation Modeling and Analysis
1 1.MLE 2.K-function & variants 3.Residual methods 4.Separable estimation 5.Separability tests Estimation & Inference for Point Processes.
1 Applications of point process modeling, separability testing, & estimation to wildfire hazard assessment 1.Background 2.Problems with existing models.
Presenting: Assaf Tzabari
1 Some Current Problems in Point Process Research: 1. Prototype point processes 2. Non-simple point processes 3. Voronoi diagrams.
Probability Distributions Continuous Random Variables.
Some standard univariate probability distributions
Lecture II-2: Probability Review
Introduction to Stochastic Models GSLM 54100
4-1 Continuous Random Variables 4-2 Probability Distributions and Probability Density Functions Figure 4-1 Density function of a loading on a long,
Exponential Distribution & Poisson Process
A Review of Probability Models
1 Exponential Distribution & Poisson Process Memorylessness & other exponential distribution properties; Poisson process and compound P.P.’s.
1 Performance Evaluation of Computer Systems By Behzad Akbari Tarbiat Modares University Spring 2009 Introduction to Probabilities: Discrete Random Variables.
Renewal processes. Interarrival times {0,T 1,T 2,..} is an i.i.d. sequence with a common distribution fct. F S i =  j=1 i T j {S i } is a nondecreasing,
Forecasting occurrences of wildfires & earthquakes using point processes with directional covariates Frederic Paik Schoenberg, UCLA Statistics Collaborators:
The Poisson Process. A stochastic process { N ( t ), t ≥ 0} is said to be a counting process if N ( t ) represents the total number of “events” that occur.
Functions of Random Variables. Methods for determining the distribution of functions of Random Variables 1.Distribution function method 2.Moment generating.
Generalized Semi-Markov Processes (GSMP)
Intro. to Stochastic Processes
Moment Generating Functions
Random Sampling, Point Estimation and Maximum Likelihood.
Dong-Yun Kim, Chao Han, and Evan Brooks Virginia Tech Department of Statistics.
Use of moment generating functions 1.Using the moment generating functions of X, Y, Z, …determine the moment generating function of W = h(X, Y, Z, …).
Continuous Distributions The Uniform distribution from a to b.
Stochastic Models Lecture 2 Poisson Processes
Continuous probability distributions
Convergence in Distribution
Week11 Parameter, Statistic and Random Samples A parameter is a number that describes the population. It is a fixed number, but in practice we do not know.
Goodness of fit testing for point processes with application to ETAS models, spatial clustering, and focal mechanisms (USGS) Frederic Paik Schoenberg,
4-1 Continuous Random Variables 4-2 Probability Distributions and Probability Density Functions Figure 4-1 Density function of a loading on a long,
1. Difficulty of point process model evaluation. 2. RELM and CSEP. 3. Numerical summaries (L-test, N-test, etc.). 4. Functional summaries (error diagrams,
CLT for Degrees of Random Directed Geometric Networks Yilun Shang Department of Mathematics, Shanghai Jiao Tong University May 18, 2008.
Generalized Semi- Markov Processes (GSMP). Summary Some Definitions The Poisson Process Properties of the Poisson Process  Interarrival times  Memoryless.
Consistency An estimator is a consistent estimator of θ, if , i.e., if
Stats Probability Theory Summary. The sample Space, S The sample space, S, for a random phenomena is the set of all possible outcomes.
Expectation. Let X denote a discrete random variable with probability function p(x) (probability density function f(x) if X is continuous) then the expected.
S TOCHASTIC M ODELS L ECTURE 2 P ART II P OISSON P ROCESSES Nan Chen MSc Program in Financial Engineering The Chinese University of Hong Kong (ShenZhen)
The generalization of Bayes for continuous densities is that we have some density f(y|  ) where y and  are vectors of data and parameters with  being.
1 Applications of space-time point processes in wildfire forecasting 1.Background 2.Problems with existing models (BI) 3.A separable point process model.
1 EE571 PART 3 Random Processes Huseyin Bilgekul Eeng571 Probability and astochastic Processes Department of Electrical and Electronic Engineering Eastern.
Geology 6600/7600 Signal Analysis 09 Sep 2015 © A.R. Lowry 2015 Last time: Signal Analysis is a set of tools used to extract information from sequences.
Lecture 2: Everything you need to know to know about point processes Outline: basic ideas homogeneous (stationary) Poisson processes Poisson distribution.
Probability and Statistics in Environmental Modeling.
Ver Chapter 5 Continuous Random Variables 1 Probability/Ch5.
Week 21 Statistical Model A statistical model for some data is a set of distributions, one of which corresponds to the true unknown distribution that produced.
4-1 Continuous Random Variables 4-2 Probability Distributions and Probability Density Functions Figure 4-1 Density function of a loading on a long,
Statistical Modelling
Engineering Probability and Statistics - SE-205 -Chap 4
水分子不時受撞,跳格子(c.p. 車行) 投骰子 (最穩定) 股票 (價格是不穏定,但報酬過程略穩定) 地震的次數 (不穩定)
Exponential Distribution & Poisson Process
Pertemuan ke-7 s/d ke-10 (minggu ke-4 dan ke-5)
Spatial Point Pattern Analysis
Rick Paik Schoenberg, UCLA Statistics
Rick Paik Schoenberg, UCLA Statistics
Example Suppose X ~ Uniform(2, 4). Let . Find .
STOCHASTIC HYDROLOGY Random Processes
Presentation transcript:

1 1.Definitions & examples 2.Conditional intensity & Papangelou intensity 3.Models a) Renewal processes b) Poisson processes c) Cluster models d) Inhibition models An Introduction to Point Processes

2 Centroids of Los Angeles County wildfires, Point process: a random point pattern. Point pattern: a collection of points in some space.

3

4

5

6 Aftershocks from global large earthquakes

7 Epicenters & times of microearthquakes in Parkfield, CA

8 Hollister, CA earthquakes: locations, times, & magnitudes Marked point process: a random variable (mark) with each point.

9 Los Angeles Wildfires: dates and sizes

10 Time series:

11

12 Time series: Palermo football rank vs. time Marked point process: Hollister earthquake times & magnitudes

13 Modern definition: A point process N is a Z + -valued random measure N(a,b) = Number of points with times between a & b. N(A) = Number of points in the set A. Antiquated definition: a point process N(t) is a right-continuous, Z + -valued stochastic process: --x x x x---x-x t T N(t) = Number of points with times < t. Problem: does not extend readily to higher dimensions.

14 More Definitions:  -finite: finite number of pts in any bounded set. Simple: N({x}) = 0 or 1 for all x, almost surely. (No overlapping pts.) Orderly: N(t, t+  )/  ----> p 0, for each t. Stationary: The joint distribution of {N(A 1 +u), …, N(A k +u)} does not depend on u. Notation & Calculus: ∫ A f(x) dN = ∑f(x i ), for x i in A. ∫ A dN = N(A) = # of points in A.

15 Intensities (rates) and Compensators x-x x x---x x t- t t+ T Consider the case where the points are observed in time only. N[t,u] = # of pts between times t and u. Overall rate:  (t) = lim  t -> 0 E{N[t, t+  t)} /  t. Conditional intensity: (t) = lim  t -> 0 E{N[t, t+  t) | H t } /  t, where H t = history of N for all times before t. If N is orderly, then (t) = lim  t -> 0 P{N[t, t+  t) > 0 | H t } /  t. Compensator: predictable process C(t) such that N-C is a martingale. If (x) exists, then ∫ o t (u) du = C(t). Papangelou intensity: p (t) = lim  t -> 0 E{N[t, t+  t) | P t } /  t, where P t = information on N for all times before and after t.

16 Intensities (rates) and Compensators x-x x x---x x t- t t+ T These definitions extend to space and space-time: Conditional intensity: (t,x) = lim  t,  x -> 0 E{N[t, t+  t) x B x,  x | H t } /  t  x, where H t = history of N for all times before t, and B x,  x is a ball around x of size  x. Compensator: ∫ A (t,x) dt dx = C(A). Papangelou intensity: p (t,x) = lim  t,  x -> 0 E{N[t, t+  t) x B x,  x | P t,x } /  t  x, where P t,x = information on N for all times and locations except (t,x).

17 Some Basic Properties of Intensities: Fact 1 (Uniqueness). If exists, then it determines the distribution of N. (Daley and Vere-Jones, 1988). Fact 2 (Existence). For any simple point process N, the compensator C exists and is unique. (Jacod, 1975) Typically we assume that exists, and use it to model N. Fact 3 (Kurtz Theorem). The avoidance probabilities, P{N(A)=0} for all measurable sets A, also uniquely determine the distribution of N. Fact 4 (Martingale Theorem). For any predictable process f(t), E ∫ f(t) dN = E ∫ f(t) (t) dt. Fact 5 (Georgii-Zessin-Nguyen Theorem). For any ex-visible process f(x), E ∫ f(x) dN = E ∫ f(x) p (x) dx.

18 Some Important Point Process Models: 1)Renewal process. The inter-event times: t 2 - t 1, t 3 - t 2, t 4 - t 3, etc. are independent and identically distributed random variables. (Classical density estimation.) Ex.: Normal, exponential, power-law, Weibull, gamma, log-normal.

19 2) Poisson process. Fact 6: If N is orderly and does not depend on the history of the process, then N is a Poisson process: N(A 1 ), N(A 2 ), …, N(A k ) are independent, and each has the Poisson dist.: P{N(A) = j} = [C(A)] j exp{-C(A)} / j!. Recall: C(A) = ∫ A (x) dx. a)Stationary (homogeneous) Poisson process: (x) = . Fact 7: Equivalent to a renewal process with exponential inter- event times. b)Inhomogeneous Poisson process: (x) = f(x), where f(x) is some fixed, deterministic function.

20 The Poisson process is the limiting distribution in many important results: Fact 8 (thinning; Westcott 1976): Suppose N is simple, stationary, & ergodic.

21 Fact 9 (superposition; Palm 1943): Suppose N is simple & stationary. Then M k --> stationary Poisson.

22 Fact 10 (translation; Vere-Jones 1968; Stone 1968): Suppose N is stationary. Then M k --> stationary Poisson. For each point x i in N, move it to x i + y i, where {y i } are iid. Let M k be the result of k such translations.

23 Fact 11 (rescaling; Meyer 1971): Suppose N is simple and has at most one point on any vertical line. Rescale the y-coordinates: move each point (x i, y i ) to (x i, ∫ o y i (x i,y) dy). Then the resulting process is stationary Poisson.

24 3) Some cluster models. a)Neyman-Scott process: clusters of points whose centers are formed from a stationary Poisson process. Typically each cluster consists of a fixed integer k of points which are placed uniformly and independently within a ball of radius r around each cluster’s center. b)Cox-Matern process: cluster sizes are random: independent and identically distributed Poisson random variables. c)Thomas process: cluster sizes are Poisson, and the points in each cluster are distributed independently and isotropically according to a Gaussian distribution. d)Hawkes (self-exciting) process: “mothers” are formed from a stationary Poisson process, and each produces a cluster of “daughter” points, and each of them produces a cluster of further “daughter” points, etc. (t, x) =  + ∑ g(t-t i, ||x-x i ||). t i < t

25 4) Some inhibition models. a)Matern (I) process: first generate points from a stationary Poisson process, and then if there are any pairs of points within distance d of each other, delete both of them. b)Matern (II) process: generate a stationary Poisson process, then index the points j = 1,2,…,n at random, and then successively delete any point j if it is within distance d from any retained point with smaller index. c) Simple Sequential Inhibition (SSI): Keep simulating points from a stationary Poisson process, deleting any if it is within distance d from any retained point, until exactly k points are kept. d)Self-correcting process: Hawkes process where g can be negative: (t, x) =  + ∑ g(t-t i, ||x-x i ||). t i < t

26 Poisson (100) Poisson (50+50x+50y) Neyman-Scott(10,5,0.05) Cox-Matern(10,5,0.05) Thomas (10,5,0.05) Matern I (200, 0.05) Matern II (200, 0.05) SSI (200, 0.05)

27 In modeling a space-time marked point process, usually directly model (t,x,a). For example, for Los Angeles County wildfires: Windspeed. Relative Humidity, Temperature, Precipitation, Tapered Pareto size distribution f, smooth spatial background . (t,x,a) =  1 exp{  2 R(t) +  3 W(t) +  4 P(t)+  5 A(t;60) +  6 T(t) +  7 [  8 - D(t)] 2 }  (x) g(a). Could also include fuel age, wind direction, interactions…

28 r = 0.16 (sq m)

29

30 (F) (sq m)

31 In modeling a space-time marked point process, usually directly model (t,x,a). For example, for Los Angeles County wildfires: Windspeed. Relative Humidity, Temperature, Precipitation, Tapered Pareto size distribution f, smooth spatial background . (t,x,a) =  1 exp{  2 R(t) +  3 W(t) +  4 P(t)+  5 A(t;60) +  6 T(t) +  7 [  8 - D(t)] 2 }  (x) g(a). Could also include fuel age, wind direction, interactions…

32

33

34 In modeling a space-time marked point process, usually directly model (t,x,a). For example, for Los Angeles County wildfires: Relative Humidity, Windspeed, Precipitation, Aggregated rainfall over previous 60 days, Temperature, Date Tapered Pareto size distribution f, smooth spatial background . (t,x,a) =  1 exp{  2 R(t) +  3 W(t) +  4 P(t)+  5 A(t;60) +  6 T(t) +  7 [  8 - D(t)] 2 }  (x) g(a). Could also include fuel age, wind direction, interactions…

35 (Ogata 1998)

36 Simulation. 1)Sequential. a) Renewal processes are easy to simulate: generate iid random variables z 1, z 2, … from the renewal distribution, and let t 1 =z 1, t 2 = z 1 + z 2, t 3 = z 1 +z 2 +z 3, etc. b) Reverse Rescaling. In general, can simulate a Poisson process with rate 1, and move each point (t i, x i ) to (t i, y i ), where x i = ∫ o y i (t i,x) dx. 2)Thinning. If m = sup (t, x), first generate a Poisson process with rate m, and then keep each point (t i, x i ) with probability (t i, x i )/m.

37 Summary: Point processes are random measures: N(A) = # of points in A. (t,x) = Expected rate around x, given history < time t. Classical models are renewal & Poisson processes. For Poisson processes, (t,x) is deterministic. Poisson processes are limits in thinning, superposition, translation, and rescaling theorems. Non-Poisson processes may have clustering (Neyman- Scott, Cox-Matern, Thomas, Hawkes) or inhibition (MaternI, MaternII, SSI, self-correcting). Next time: How to estimate the parameters in these models, and how to tell how well a model fits….