Distributions Reference (Pdf)

Distributions Reference (Pdf)

MLEtools.nb 1 Reference: some basic distributions An excellent reference for probability distributions is: Evans, Merran, Nicholas Hastings, and Brian Peacock. 2000. Statistical Distributions, 3rd Edition. John Wiley & Sons. New York. Continuous Expected, or mean E Z = zf z „ z Variance VAR Z = E Z - E Z 2 = z - E z 2 f z „ z @ D Ÿ H L For continuous variables, the probability density function is the probability of the value z given the parameters H L @H @ DL D Ÿ H @ DL H L ü Uniform Uniform Distribution: a uniform distribution on the interval [a,b] where a < b probability density function: 1 f z = ÅÅÅÅÅÅÅÅÅÅb-a , where a § z § b b+a mean: ÅÅÅÅÅÅÅÅÅÅ2 b-a 2 variance: ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ12 H L H L ü r dunif z, min = a, max = b # generate 1 uniform random number over the range 1.4 to 2.4 runif H1, 1.4, 2.4 L ü mathematicaH L 1 f z_, a_, b_ := b − a ∗ random number over range 0,1 ∗ @ D Random H0.0407265 L @D ∗ random number of specified range ∗ H L MLEtools.nb 2 Random Real, 3, 6 4.87348 @ 8 <D ü Normal Normal Distribution: has two parameters: m is the location parameter, in this case the mean; s is the scale parameter, in this case the standard deviation probability density function: 2 f z = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ1 exp - ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅz-m 2 ps2 2 s2 mean: m H L variance: s2 è!!!!!!!!!!! !!! H L I M note: plotting f z will give you the familiar bell curve ü r H L probability density function dnorm z, mean, sd # plot the pdf curve dnorm x, 4, 1 ,0,7 H L random number generation, where n is the number random numbers to generation. In other words, n = 1 will create 1 random number. H H L L rnorm n, mean, sd ü mathematicaH L 1 z − m 2 f z_, m_, σ_ := Exp − 2 2 π σ2 2 σ Plot f z, 4, 1 , z, 0, 7 ; H L @ D è!!!!!!!!!!!!! A E 0.4 @ @ D 8 <D 0.3 0.2 0.1 1 2 3 4 5 6 7 MLEtools.nb 3 Show that pdf integrates to 1 Integrate f z, 1, 1 , z, −∞, ∞ 1 @ @ D 8 <D << Statistics`ContinuousDistributions` ∗ normal random number with mean of 3 and standard deviation 2 ∗ Random NormalDistribution 3, 2 H1.87451 L @ @ DD ü Lognormal Lognormal Distribution: has two parameters: m is the location parameter; s is the shape parameter probability density function: 1 ln z -log m 2 f z = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ exp - ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ2 , where z > 0, m >0, and s >0 z 2 ps2 2 s 2 mean: m exp ÅÅÅÅÅÅÅs 2 H H L H LL median: m è!!!!!!!!!!! !!! H L I M s2 is the variance of the ln(z) variance: m2I expMs2 exp s2 - 1 note: This describes a variable in which the logarithm of the variable is a normal distribution. Whereas the normal distribu- tion is additive, the lognormal distribution is multiplicative. H L H H L L ü r dlnorm z, meanlog = m, sdlog =σ R uses a slightly different form of the lognormal model. Use the following form so that m is defined in the same way as described in the f(zH) equation above. L dlnorm z, log m , σ # generate random number rlnorm H1, meanlogH L = 1,L sdlog = 1.2 ü mathematica H L 1 Log z − Log m 2 f z_, m_, σ_ := Exp − 2 z 2 π σ2 2 σ H @ D @ DL @ D è!!!!!!!!!!!!! A E MLEtools.nb 4 Plot f z, 1.7, 1 , z, 0, 20 ; @ @ D 8 <D 0.3 0.2 0.1 5 10 15 20 Show that pdf integrates to 1 Integrate f z, 1, 1 , z, 0, ∞ 1 @ @ D 8 <D Derive expected value Integrate zf z, m, σ , z, 0, ∞ , Assumptions → σ>0, m > 0, σ∈Reals, m ∈ Reals 2 σ 2 m @ @ D 8 < 8 <D Derive variance Integrate z − Integrate zf z, m, σ , z, 0, ∞ , Assumptions → σ>0, m > 0, σ∈Reals, m ∈ Reals 2 f z, m, σ , z, 0, ∞ , Assumptions → σ>0, m > 0, σ∈Reals, m ∈ Reals @ 2 2 Hσ −1 +σ m2 @ @ D 8 < 8 <DL @ D 8 < 8 <D << Statistics`ContinuousDistributions` ∗ IlognormalM random number with the first number being the Log m and the second number being σ. In the case below, m = 3 and σ=2 ∗ Random LogNormalDistribution Log 3 ,2 H @ D 6.94065 L @ @ @ D DD MLEtools.nb 5 ü Exponential Exponential Distribution: has three parameters: m is the location parameter; s is the shape parameter probability density function: f z = a Exp -az, where z > 0 1 mean: ÅÅÅÅa 1 variance: ÅÅÅÅÅÅÅa2 note: ContinuousH L distribution@ D analog of Poisson. Can be used for dispersal distance. Widely used in Baysian Analysis ü r dexp z, rate = a # generate random number rexp H1, rate = aL ü mathematicaH L f z_, a_ := a Exp −az Plot f z, 0.5 , z, 0, 15 , PlotRange → All ; @ D @ D 0.5 @ @ D 8 < D 0.4 0.3 0.2 0.1 2 4 6 8 10 12 14 << Statistics`ContinuousDistributions` ∗ exponential random number with a = 0.5 ∗ Random ExponentialDistribution 0.5 H2.1514 L @ @ DD MLEtools.nb 6 ü Gamma Gamma Distribution: has two parameters: a is the rate parameter; n is the shape parameter probability density function: an n-1 f z = ÅÅÅÅÅÅÅÅÅÅÅG n Exp -az z , where z > 0 n mean: ÅÅÅÅa n variance: ÅÅÅÅÅÅÅ2 H L H aL @ D note: can have a long tail. When n = 1, the gamma distribution becomes the exponential distribution. When n = degrees of freedom / 2 and a = 2, the gamma distribution becomes the chi-square distribution. ü r dgamma z, shape = n , rate = a # generate random number rgamma H1, shape = n, rate = a L ü mathematica H L an f z_, a_, n_ := Exp −az zn−1 Gamma n Plot f z, 1, 1 , z, 0, 12 , PlotRange → All ; @ D @ D 1 @ D @ @ D 8 < D 0.8 0.6 0.4 0.2 2 4 6 8 10 12 MLEtools.nb 7 Plot f z, 1, 2 , z, 0, 12 , PlotRange → All ; 0.35 @ @ D 8 < D 0.3 0.25 0.2 0.15 0.1 0.05 2 4 6 8 10 12 Plot f z, 0.5, 2 , z, 0, 12 , PlotRange → All ; 0.175 @ @ D 8 < D 0.15 0.125 0.1 0.075 0.05 0.025 2 4 6 8 10 12 Derive expected value Integrate zf z, a, n , z, 0, ∞ , Assumptions → a > 0, n > 0, a ∈ Reals, n ∈ Reals n a @ @ D 8 < 8 <D Derive variance Integrate z − Integrate zf z, a, n , z, 0, ∞ , Assumptions → a > 0, n > 0, a ∈ Reals, n ∈ Reals 2 f z, a, n , z, 0, ∞ , Assumptions → a > 0, n > 0, a ∈ Reals, n ∈ Reals @ n H @ @ D 8 < 8 <DL a2 @ D 8 < 8 <D << Statistics`ContinuousDistributions` ∗ gamma random number with n = 2 and a = 0.5 ∗ Random GammaDistribution 2.0 , 1 0.5 H2.82494 L @ @ ê DD MLEtools.nb 8 ü Chi-Square Note: this section is a bit incomplete Chi-Square Distribution: A squared normal distribution is a chi-square distribution. probability density function for 1 degree of freedom: f z = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ1 ÅÅÅÅÅÅÅÅ exp - ÅÅÅÅÅÅÅÅÅÅÅz 2 ps2 z 2 s2 probability densityè!!!!!!! function!!!! !!!!!! for n degrees of freedom: H L 1 Hn 2 -1 L z f z = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅn n 2ÅÅÅÅÅÅÅÅÅÅÅÅÅÅn z exp - ÅÅÅÅÅÅÅÅÅÅÅ2 , for z ¥ 0 s 2 G ÅÅÅÅ2 2 s mean: n s2 H ê L 4 H ê L variance: 2 n s H L H L H L notes: - n can also be thought of as a shape parameter in the context of a general distribution - often assume unit variance, s = 1 1 - G ÅÅÅÅ2 = p è!!!! ü rH L dchisq ü mathematica HL 1 z f z_, σ_ := Exp − 2 2 π σ2 z 2 σ Plot f z, 2.1 , z, 0, 7 ; @ D è!!!!!!!!!!!!!!!!! A E 1.75 1.5 @ @ D 8 <D 1.25 1 0.75 0.5 0.25 1 2 3 4 5 6 7 Show that pdf integrates to 1 MLEtools.nb 9 Integrate f z, 2.1 , z, 0, ∞ 1. @ @ D 8 <D << Statistics`ContinuousDistributions` ∗ Chi−Square distributed random number with 1 degree of freedom ∗ Random ChiSquareDistribution 1 H0.111756 L @ @ DD Add non-centrality, or location, parameter with a single degree of freedom, although no scale parameter mean: 1 + m Two forms are given below −1 z +μ 1 z 4 f z_, μ_ := Exp − BesselI −1 2, μ z 2 2 μ 1 1 1 è!!!!!!! z +μ f@z_, μ_D := Hypergeometric0F1A E J N , z μ A ê ExpE− 2 4 2 π z 2 NIntegrate f z, 2.0 , z, 0, 100000 @ D A E è!!!!!!!!!!!! A E 1. @ @ D 8 <D NIntegrate zf z, 2.0 , z, 0, 100000 3. @ @ D 8 <D Integrate zf z, μ , z, 0, ∞ , Assumptions → μ>0, μ∈Reals 1 +μ @ @ D 8 < 8 <D 1 1 Hypergeometric0F1 , z μ 2 4 Cosh z μ A E 1 1 z +μ è!!!!!!! μ + − μ − ExpA zE Exp z Exp 2 2 π z 2 è!!!!!!! è!!!!!!! 1 z +μ ExpI Az μ +EExp − A z μ EMè!!!!!!!!!!!! Exp −A E 2 2 π z 2 1è!!!!!!! è!!!!!!!z +μ 1 z +μ IA ExpE zAμ−EM +è!!!!!!!!!!!! AExp − zEμ− 2 2 π z 2 2 2 π z 2 è!!!!!!! è!!!!!!! è!!!!!!!!!!!! A E è!!!!!!!!!!!! A E MLEtools.nb 10 ü Bimodal Normal Normal Distribution: has two parameters: m is the location parameter, in this case the mean; s is the scale parameter, in this case the standard deviation probability density function: 2 f z = ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅ1 exp - ÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅÅz-m 2 ps2 2 s2 mean: m H L variance: s2 è!!!!!!!!!!! !!! H L I M note: plotting f z will give you the familiar bell curve Here I extend that to a bimodal normal distribution.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    26 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us