Date Tuesday January 22,2019 COMPUTATIONAL STATISTICS MSDS 628 ol Tif James Wilson re sampling methods boostrap MC MC kernel density estimation histogram artificial neural networks additivemodels Bayesian truth serum NLP latent Dirichlet Allocation Overall topic w distribution Experimental design multi armed bandit gambling Reinforcement learning Review of Statistical Models Random Variable Number w prob distr Whatvalues with what probability Definition X possible values for a random variable X Pdf IP Xix XC X coff i IP XIX X C X Example 3 coin head ii iii is Iii 3 Is
3
Embed
COMPUTATIONAL STATISTICS - Project Jupyter...Date Tuesday January 22,2019 COMPUTATIONAL STATISTICS MSDS 628 ol Tif James Wilson resamplingmethods boostrap MCMC kernel densityestimation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Date Tuesday January 22,2019
COMPUTATIONAL STATISTICSMSDS 628 ol TifJames Wilson
Bayesian truth serumNLP latent Dirichlet Allocation Overalltopic wdistributionExperimental design multi armed bandit gamblingReinforcement learning
Review of StatisticalModelsRandomVariable Number w probdistr
Whatvalues with whatprobabilityDefinition
X possiblevalues for a randomvariable XPdf IP Xix XC Xcoff i IP XIX XC X
Example3coin head
ii iii is Iii3 Is
Example Icontinue RVsX n U OilR X O 5 O
Xcould take infinitevalue1 tooReason why we think about intervals forcontinousvariables
IPCXEO 5 42 PDF
For continous RVs we use 1PMcA contained insome set A
PHEB SB fCxDX
ExampleIPCYEO 8
f I IEirIIF a Fca flat
Distributions
Expected value Standard deviationLinear
varianceQuadratic
Jointly distributedrandomvariablesFla b REX Ea YEb La b too
If youhave multiple rvs knowingtheir jointdistribution gives all information about the RV'smarginaldistribution Also havingthe joint Pdf is equivalentto havingtheconditionaldistribution ofeachRV giventhe restHavingthemarginalPdfs ofeach RVsdoes not giveusthe joint distribution or conditional distributions ofall RVs