Spatial Statistics (SGG 2413) Probability Distribution Assoc. Prof. Dr. Abdul Hamid b. Hj. Mar Iman Former Director Centre for Real Estate Studies Faculty of Engineering and Geoinformation Science Universiti Tekbnologi Malaysia Skudai, Johor SGG2413 - Theory of Probability 1 Learning Objectives • Overall: To expose students to the concepts of probability • Specific: Students will be able to: * define what are probability and random variables * explain types of probability * write the operational rules in probability * understand and explain the concepts of probability distribution SGG2413 - Theory of Probability 2 Contents • • • • Basic probability theory Random variables Addition and multiplication rules of probability Discrete probability distribution: Binomial probability distribution, Poisson probability distribution • Continuous probability distribution • Normal distribution and standard normal distribution • Joint probability distribution SGG2413 - Theory of Probability 3 Basic probability theory • Probability theory examines the properties of random variables, using the ideas of random variables, probability & probability distributions. • Statistical measurement theory (and practice) uses probability theory to answer concrete questions about accuracy limits, whether two samples belong to the same population, etc. • probability theory is central to statistical analyses SGG2413 - Theory of Probability 4 Basic probability theory • Vital for understanding and predicting spatial patterns, spatial processes and relationships between spatial patterns • Essential in inferential statistics: tests of hypotheses are based on probabilities • Essential in the deterministic and probabilistic processes in geography: describe real world processes that produce physical or cultural patterns on our landscape SGG2413 - Theory of Probability 5 Basic probability theory (cont.) • Deterministic process – an outcome that can be predicted almost with 100% certainty. • E.g. some physical processes: speed of comet fall, travel time of a tornado, shuttle speed • Probabilistic process – an outcome that cannot be predicted with a 100% certainty • Most geographic situations fall into this category due to their complex nature • E.g. floods, draught, tsunami, hurricane • Both categories of process is based on random variable concept SGG2413 - Theory of Probability 6 Basic probability theory (cont.) • Random probabilistic process – all outcomes of a process have equal chance of occurring. E.g. * Drawing a card from a deck, rolling a die, tossing a coin …maximum uncertainty • Stochastic processes – the likelihood of a particular outcome can be estimated. From totally random to totally deterministic. E.g. * Probability of floods hitting Johor: December vs. January …probability is estimated based on knowledge which will affect the outcome SGG2413 - Theory of Probability 7 Random Variables • Definition: – A function of changeable and measurable characteristic, X, which assigns a real number X(ζ) to each outcome ζ in the sample space of a random experiment • Types of random variables: – Continuous. E.g. income, age, speed, distance, etc. – Discrete. E.g. race, sex, religion, etc. S SGG2413 - Theory of Probability ζ X(ζ) = x x Sx 8 Basic concepts of random variables • Sample Point – The outcome of a random experiment • Sample Space, S – The set of all possible outcomes – Discrete or continuous • Events – A set of outcomes, thus a subset of S – Certain, Impossible and Elementary SGG2413 - Theory of Probability 9 Basic concepts of random variables (cont.) • E.g. rolling a dice… Space…S = {1, 2, 3, 4, 5, 6} Event…Odd numbers: A = {1, 3, 5} …Even numbers: B = {2,4,6} Sample point…1, 2,.. • Let S be a sample space of an experiment with a finite or countable number of outcomes. • We assign p(s) to each outcome s. • We require that two conditions be met: 0 p 1 for each s S. sS p(s) = 1 SGG2413 - Theory of Probability 10 Basic concepts of random variables (cont.) E.g. rolling a dice… Outcome, x Prob. x Cumulative prob. X 1 1/6 1/6 = 0.166 2 1/6 2/6 = 0.333 3 1/6 3/6 = 0.500 4 1/6 4/6 = 0.666 5 1/6 5/6 = 0.833 6 1/6 6/6 = 1.000 SGG2413 - Theory of Probability 11 Types of Random Variables • Continuous – Probability Density Function Marginal change: • Discrete – Probability Mass Function No marginal change: dFX x fX x dx PX xk P X xk Bounded area: No bounded area: FX x x f X t dt FX x PX xk u x xk k SGG2413 - Theory of Probability 12 Types of Random Variables - continuous fX(x) fX(x) dx x P x X x dx f X x dx SGG2413 - Theory of Probability 13 Probability: Law of Addition If A and B are not mutually exclusive events: P(A or B) = P(A) + P(B) – P(A and B) E.g. What is the probability of types of coleoptera found on plant A or plant B? Plant A P(A or B) = P(A) + P(B) – P(A and B) Plant B = 5/10 + 3/10 – 2/10 5 2 3 = 6/10 = 0.6 Types of plant coleoptera SGG2413 - Theory of Probability 14 Probability: Law of Addition (cont.) If A and B are mutually exclusive events: P(A or B) = P(A) + P(B) E.g. What is the probability of types of coleoptera found on plant A or plant B? Plant A Plant B P(A or B) = P(A) + P(B) = 5/10 + 3/10 5 3 = 8/10 = 0.8 2 Types of plant coleoptera SGG2413 - Theory of Probability 15 Probability: Law of Multiplication If A and B are statistically dependent, the probability that A and B occur together: P(A and B) = P(A) P(B|A) where P(B|A) is the probability of B conditioned on A. If A and B are statistically independent: P(B|A) = P(B) and then P(A and B) = P(A) P(B) SGG2413 - Theory of Probability 16 P(A|B) Plant A 5 2 Plant A Plant B Plant B 5 3 3 2 Types of plant coleoptera Types of plant coleoptera A & B Statistically dependent: A & B Statistically independent: P(A and B) = P(A) P(B|A) = (5/10)(2/10) = 0.5 x 0.2 = 0.1 P(A and B) = P(A) P(B) = (5/10)(3/10) = 0.5 x 0.3 = 0.15 SGG2413 - Theory of Probability 17 Discrete probability distribution • Let’s define x = no. of bedroom of sampled houses • Let’s x = {2, 3, 4, 5} • Also, let’s probability of each outcome be: nx 20 40 30 10 100 P(x) 0.2 0.4 0.3 0.1 1.0 0.45 Probability, p(x) X 2 3 4 5 Total 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 2 SGG2413 - Theory of Probability 3 4 5 No. of bedroom, x 18 Expected Value and Variance • The expected value or mean of X is E X tf X t dt continuous E X xk PX xk discrete k • Properties • The variance of X is 2 Var X 2 E X E X • The standard deviation of X is Std X Var X • Properties E c c Var c 0 E cX cE X Var cX c2Var X E X c E X c Var X c Var X SGG2413 - Theory of Probability 19 More on Mean and Variance • Physical Meaning – If pmf is a set of point masses, then the expected value μ is the center of mass, and the standard deviation σ is a measure of how far values of x are likely to depart from μ • Markov’s Inequality P X a EX a • Chebyshev’s Inequality 2 P X a 2 a 1 P X k 2 k • Both provide crude upper bounds for certain r.v.’s but might be useful when little is known for the r.v. SGG2413 - Theory of Probability 20 Discrete probability distribution – Maduria magniplaga Tree No. of fruit landings (Xi) No. of fruits with borers attack (fXi) Prob. of fruit landings Expected no. fruits with borers (pXi = fXi/Xi) (fXi x pXi) (fXi –mean)2 (fXi –mean)2 x pXi 1 20 8 0.13 1.05 1.42 0.18 2 10 6 0.07 0.39 0.66 0.04 3 15 4 0.10 0.39 7.90 0.77 4 14 6 0.09 0.55 0.66 0.06 5 24 8 0.16 1.25 1.42 0.22 6 20 12 0.13 1.57 26.93 3.52 7 18 9 0.12 1.06 4.79 0.56 8 10 4 0.07 0.26 7.90 0.52 9 14 2 0.09 0.18 23.14 2.12 10 8 2 0.05 0.10 23.14 1.21 Sum 153 61 1.00 Mean 6.81 Variance 9.21 Std. dev. 3.04 SGG2413 - Theory of Probability 21 Discrete probability distribution – Maduria magniplaga • Expected no. of fruits with borers: E(Xi) = X.px = (fXi.Xi/Xi) = 6.81 ≈7 • Variance of fruit borers’ attack: 2 = E[(X-E(X))2] = (fni – mean)2 x pXi = 9.21 ● Standard deviation of fruit borers’ attack: = 9.21 = 3.04 SGG2413 - Theory of Probability 22 Discrete probability distribution: Binomial • Outcomes come from fixed n random occurrences, X • Occurrences are independent of each other • Has only two outcomes, e.g. ‘success’ or • ‘failure’ • The probability of "success" p is the same for each occurrence • X has a binomial distribution with parameters n and p, abbreviated X ~ B(n, p). SGG2413 - Theory of Probability 23 Discrete probability distribution: Binomial (cont.) The probability that a random variable X ~ B(n, p) is equal to the value k, where k = 0, 1,…, n is given by where Mean and variance: SGG2413 - Theory of Probability 24 Discrete probability distribution: Binomial (cont.) • E.g. The Road Safety Department discovered that the number of potential accidents at a road stretch was 18, of which 4 are fatal accidents. Calculate the mean and variance of the non-fatal accidents. • = np = 18 x 0.78 = 14 • 2 = np(1-p) = 14 x (1-0.78) = 3.08 SGG2413 - Theory of Probability 25 Cumulative Distribution Function • Defined as the probability of the event {X≤x} • Properties Fx(x) FX x P X x 1 x 0 FX x 1 Fx(x) lim FX x 1 1 x ¾ lim FX x 0 x if a b then FX a FX a P a X b FX b FX a ½ ¼ 0 1 2 3 x P X x 1 FX x SGG2413 - Theory of Probability 26 Probability Density Function dF x • The pdf is computed from f X x X dx • Properties P a X b f X x dx b a FX x x fX(x) f X t dt 1 f X t dt fX(x) dx • For discrete r.v f X x PX xk x xk x P x X x dx f X x dx k SGG2413 - Theory of Probability 27 Conditional Distribution • The conditional distribution function of X given the event B P X x B FX x | B P B • The conditional dFX x |pdf B is fX x | B • The distribution function can be written as a weighted sum of conditional distribution functions n FX x | B FX x | Ai P Ai i 1 dx where Ai mutally exclusive and exhaustive events SGG2413 - Theory of Probability 28 Joint Distributions • Joint Probability Mass Function of X, Y p XY x j , yk P X x j Y y j P X x j , Y yk • Probability of event A • Joint CMF of X, Y FXY x1 , y1 P X x1 , Y y1 • Marginal CMFs FX x FXY x, P X x FY y FXY , y P Y y PXY X , Y A pXY x j , yk jA kA • Marginal PMFs (events involving each rv in isolation) p XY x j P X x j p XY x j , yk k 1 SGG2413 - Theory of Probability 29 Conditional Probability and Expectation • The conditional CDF of Y given the event {X=x} is FY y | x y f XY x, y ' dy ' fX x • The conditional expectation of Y given X=x is E Y | x yfY y | x dy • The conditional PDF of Y given the event {X=x} is fY Y | x f XY x, y fX x f X x | y fY y fY y | x fX x SGG2413 - Theory of Probability 30 Independence of two Random Variables • X and Y are independent if {X ≤ x} and {Y ≤ y} are independent for every combination of x, y • Conditional Probability of independent R.V.s f XY x, y f X x fY y FXY x, y FX x FY y fY y | x fY y f XY x, y f X x fY y f X x | y f X x SGG2413 - Theory of Probability 31 Thank you SGG2413 - Theory of Probability 32