pset1_solutions added

.Rhistory

@@ -0,0 +1,292 @@
+df = data.frame("x" = c(1), "y" =c(2))
+df
+linearmodel = lm(y ~ x, data = df)
+linearmodel
+dunif(x, min = 0, max = 1, log = FALSE)
+matrix(0, 1, 1000)
+matrix(0, 1, 1000)
+dunif(x, min = 0, max = 1, log = FALSE)
+1000)
+dunif(x, min = 0, max = 1, log = FALSE)
+dunif(x, min = 0, max = 1, log = FALSE)
+x=matrix(0, 1, 1000)
+dunif(x, min = 0, max = 1, log = FALSE)
+x=dunif(x, min = 0, max = 1, log = FALSE)
+x
+x=punif(q, min = 0, max = 1, lower.tail = TRUE, log.p = FALSE)
+x=punif(x, min = 0, max = 1, lower.tail = TRUE, log.p = FALSE)
+x
+u <- runif(1000)
+u
+u <- runif(x)
+u
+u <-runif(x, min = 0, max = 1)
+u
+clear
+u <- runif(1000)
+u
+attributes(u)
+summarize(u)
+describe.u
+describe(u)
+x=matrix(0,1000)
+x
+describe(u)
+describe.vector(u)
+describe.matrix(u)
+str(u)
+typeof(u)
+typeof(u)
+u
+typeof(u)
+bidders1 <- u[1:500]
+bidders1 = u[1:500]
+u <- runif(x)
+x=matrix(0,1000)
+u <- runif(x)
+bidders1 <- u[1:500]
+bidders1
+x=matrix(0,1000)
+u <- runif(x)
+bidders1 <- u[1:500]
+dim(bidders1) <- c(125,4)
+bidders1`
+x=matrix(0,1000)
+u <- runif(x)
+bidders1 <- u[1:500]
+dim(bidders1) <- c(125,4)
+bidders1
+dim(bids) <- c(125,4)
+dim(bids1) <- c(125,4)
+bidders <- runif(matrix(0,1000))
+bids1 <- (3/4)*u[1:500]
+dim(bids1) <- c(125,4)
+bids1
+bidders <- runif(matrix(0,1000))
+bids1 <- (3/4)*u[1:500]
+dim(bids1) <- c(125,4)
+bids2 <- (4/5)*u[501:1000]
+dim(bids2) <- c(100,5)
+bids2
+# library(TTR)
+# djia = getYahooData("^DJI", start=20060420, end=20160420, freq="daily")
+library(xts)
+djiar = diff(log(djia$Close))[-1] # approximate returns
+plot(djiar, main="DJIA Returns", type="n")
+lines(djiar)
+library(TTR)
+djia = getYahooData("^DJI", start=20060420, end=20160420, freq="daily")
+library(xts)
+djiar = diff(log(djia$Close))[-1] # approximate returns
+plot(djiar, main="DJIA Returns", type="n")
+lines(djiar)
+library(TTR)
+djiar = getYahooData("^DJI", start=20060420, end=20160420, freq="daily")
+library(xts)
+djiar = diff(log(djia$Close))[-1] # approximate returns
+plot(djiar, main="DJIA Returns", type="n")
+lines(djiar)
+install.packages("astsa")
+library(astsa)
+.First <- function(){library(astsa)}
+1:5
+seq(1, 10, by=2)
+x=c(1:20); y=c(1:5); x*y
+ls()
+ls(pattern=2)
+ls(pattern=bid)
+ls(pattern="bid")
+ls(pattern=.4020)
+ls(pattern=.402)
+ls()
+ls(pattern=u)
+ls(pattern="u")
+help(read.table)
+x = 1:3; y = 4:6
+(u = c(x, y)) # an R vector
+(u1 = cbind(x, y))
+(u1 = cbind(t(x), t(y))
+(u1 = cbind(t(x), t(y))
+library(TTR)
+djiar = getYahooData("^DJI", start=20060420, end=20160420, freq="daily")
+library(xts)
+djiar = diff(log(djia$Close))[-1] # approximate returns
+plot(djiar, main="DJIA Returns", type="n")
+lines(djiar)
+library(tidyverse)
+library(readr)
+library(cubature)
+setwd("C:/Users/Joe/github/empirical-methods")
+dt <- read.delim("ProblemSets/pset1/bids1.csv")
+dt <- rbind(2.9679,dt) # first observation was read as a column name, so put it back as an observation
+names(dt)<- "bids"
+mean(dt$bids)
+X <- dt$bids
+I <- 1:300 # index vector
+H <- seq(from=.01, to=.3, by=.001) # bandwidth vector
+# you need full sample density estimate to do CV
+epan <- density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+# test on X[i]
+y_j <- function(j) density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(sapply(I, MSEi))
+y_j <- function(j) density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(sapply(I, MSEi))
+J <- I # another index vector
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+y_j <- function(j) density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i,j) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+y_j <- function(j) density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i,j) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+MSEi(1)
+MSEi(i=1,j=1)
+y_j(1)
+?density
+y_j(0)
+# test on X[i]
+y_j <- function(j) density(dt$bids, n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i,j) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+y_j(1)
+y_j(2)
+y_j(0)
+y_j(-1)
+y_j <- function(j) density(dt$bids,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi <- function(i,j) (y_j(j)- yhatj(i,j))^2
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+y_j(1)
+y_j(10)
+y_j(100)
+epan <- density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+epan(1)
+epan(1)$y
+y_j <- function(j) density(X,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw="SJ")$y
+MSEi_summand <- function(i,j) (y_j(j)- yhatj(i,j))^2
+MSEi_summand(1)
+MSEi_summand(i=1,j=1)
+MSEi_summand(i=1,j=2)
+MSEi_summand(i=1,j=3)
+MSEi_summand(i=1,j=4)
+MSEi_summand(i=1,j=5)
+?Vectorize
+MSEi <- sum(Vectorize(MSEi_summand))
+MSEi <- sum(Vectorize(FUN=MSEi_summand))
+MSEi <- Vectorize(FUN=MSEi_summand)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi_vec(i,J)
+?rep
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi_vec(i_vec,J)
+y_j <- function(j) density(X,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j,h) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw=h)$y
+MSEi_summand <- function(i,j,h) (y_j(j)- yhatj(i,j,h))^2
+h_vec <- rep(.1, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- sum(MSEi_vec)
+MSEhat <- sum(mapply(I,J, FUN=MSEi))
+y_j <- function(j) density(X,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j,h) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw=h)$y
+MSEi_summand <- function(i,j,h) (y_j(j)- yhatj(i,j,h))^2
+h_vec <- rep(.1, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- sum(MSEi_vec)
+epan <- density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+# test on X[i]
+y_j <- function(j) density(X,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j,h) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw=h)$y
+MSEi_summand <- function(i,j,h) (y_j(j)- yhatj(i,j,h))^2
+h_vec <- rep(.1, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi_vec(i_vec,J,h_vec)
+MSEi <- sum(MSEi_vec)
+MSEi <- sum(MSEi_vec(i_vec,h_vec,J))
+MSEi <- sum(MSEi_vec(i_vec,J,h_vec))
+MSEi
+h_vec <- rep(.2, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- sum(MSEi_vec(i_vec,J,h_vec))
+MSEi
+h_vec <- rep(.01, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- sum(MSEi_vec(i_vec,J,h_vec))
+MSEi
+h_vec <- rep(.5, time=300)
+i_vec <- rep(1, time=300)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- sum(MSEi_vec(i_vec,J,h_vec))
+MSEi
+MSEhat <- function(h){
+h_vector <- rep.int(h, 300)
+i_vector <- rep.int(i, 300)
+return(sum(mapply(i_vector, J, h_vector, FUN=MSEi)))
+}
+MSEhat <- function(h){
+h_vector <- rep.int(h, 300)
+i_vector <- rep.int(i, 300)
+return(MSEi(i_vector, J, h_vector, FUN=MSEi))
+}
+MSEhat(.1)
+# you need full sample density estimate to do CV
+epan <- density(dt$bids, kernel= "epanechnikov", bw = "SJ", adjust = 1)
+# test on X[i]
+y_j <- function(j) density(X,from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw = "SJ", adjust = 1)$y
+yhatj <- function(i,j,h) density(X[-i],from=X[j],to=X[j],n=1, kernel= "epanechnikov", bw=h)$y
+MSEi_summand <- function(i,j,h) (y_j(j)- yhatj(i,j,h))^2
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- function(i,h){
+i_vec <- rep(1, time=300)
+h_vec <- rep(.5, time=300)
+MSEi_vec(i_vec, J, h_vec, FUN=MSEi_vec)
+}
+MSEi(1,1,)
+MSEi(1,1)
+MSEi_vec <- Vectorize(FUN=MSEi_summand)
+MSEi <- function(i,h){
+i_vec <- rep(1, time=300)
+h_vec <- rep(.5, time=300)
+return(MSEi_vec(i_vec, J, h_vec))
+}
+MSEi(1,1)
+MSEhat_i <- function(I) {
+h_vec <- rep(.5, time=300)
+return(sapply(I,h_vec,FUN=MSEi))
+}
+MSEhat_i <- function(h) {
+h_vec <- rep(.5, time=300)
+return(sapply(I,h_vec,FUN=MSEi))
+}
+MSEhat_i(.1)
+MSEhat_i(.1)
+MSEhat_i(.1)
+MSEi(1,.1)
+MSEhat_i <- function(h) {
+h_vec <- rep(h, time=300)
+return(sapply(I,h_vec,FUN=MSEi))
+}
+MSEhat_i(.1)
+MSEi <- function(i,h){
+i_vec <- rep(i, time=300)
+h_vec <- rep(h, time=300)
+return(MSEi_vec(i_vec, J, h_vec))
+}
+MSEhat_i <- function(h) {
+h_vec <- rep(h, time=300)
+return(sapply(I,h_vec,FUN=MSEi))
+}
+MSEhat_i(.1)

Part 2 - Model_Selection/model-selection.log

@@ -1,3 +1,36 @@
+<<<<<<< HEAD
+This is pdfTeX, Version 3.14159265-2.6-1.40.18 (MiKTeX 2.9.6500 64-bit) (preloaded format=pdflatex 2018.2.6)  5 FEB 2020 21:55
+entering extended mode
+**./model-selection.tex
+(model-selection.tex
+LaTeX2e <2017-04-15>
+Babel <3.15> and hyphenation patterns for 75 language(s) loaded.
+(../slideHeader.tex
+
+! LaTeX Error: File `beamer.cls' not found.
+
+Type X to quit or <RETURN> to proceed,
+or enter new name. (Default extension: cls)
+
+Enter file name: 
+! Emergency stop.
+<read *> 
+
+l.3 \usefonttheme
+                 [onlymath]{serif}
+*** (cannot \read from terminal in nonstop modes)
+
+
+Here is how much of TeX's memory you used:
+ 18 strings out of 493312
+ 443 string characters out of 3138906
+ 53761 words of memory out of 3000000
+ 3659 multiletter control sequences out of 15000+200000
+ 3640 words of font info for 14 fonts, out of 3000000 for 9000
+ 1141 hyphenation exceptions out of 8191
+ 12i,0n,7p,125b,8s stack positions out of 5000i,500n,10000p,200000b,50000s
+!  ==> Fatal error occurred, no output PDF file produced!
+=======
 This is pdfTeX, Version 3.14159265-2.6-1.40.19 (TeX Live 2018) (preloaded format=pdflatex 2018.4.16)  4 FEB 2020 13:31
 entering extended mode
  restricted \write18 enabled.
@@ -1837,3 +1870,4 @@ PDF statistics:
  188 named destinations out of 1000 (max. 500000)
  280 words of extra memory for PDF output out of 10000 (max. 10000000)
 
+>>>>>>> upstream/master