Lecture 3

Third lecture proposed exercises

Lecture 2/.Rhistory

@@ -0,0 +1,360 @@
+library( RNetCDF )
+library( ncdf )
+pwd
+help?
+directory
+setwd("/Users/samuel/")
+3+3
+3+3
+# this is just an example
+pi
+? help
+? mean
+whiteside
+library(MASS)
+whiteside
+top(whiteside)
+head(whiteside)
+names( whiteside)
+summary( whiteside)
+summary( whiteside[,1])
+1970:1970
+1970:1979
+array = c( 1970:1979)
+array
+# create a vector
+x = 1:100
+y = sin(x)
+plot( x ~ y)
+plot( y ~ x)
+plot( y ~ x, line)
+? plot
+plot( y ~ x, "l")
+plot( y ~ x, type="l")
+x = 1:1000
+y = sin(x)
+plot( x ~ y , type="l")
+plot( y ~ x , type="l")
+plot( y ~ x , type="p")
+plot( y ~ x , type="")
+plot( y ~ x , type="l")
+plot( y ~ x , type="p")
+help(array)
+source("http://bioconductor.org/biocLite.R")
+?BiocUpgrade
+biocLite("BiocUpgrade")
+y
+y
+install.packages("biomaRt")
+biocLite()
+biocLite("biomaRt")
+choosebank()
+library("seqinr")
+choosebank()
+choosebank("genbank")
+query("BRCA1", "SP=Homo sapiens AND K=BRCA1")
+library("seqinr")
+choosebank("genbank")
+query("BRCA1", "SP=Homo sapiens AND K=BRCA1")
+attributes(BRCA1)
+BRCA1$req
+query("BRCA1", "SP=Homo sapiens AND AC=U61268")
+myseq <- getSequence(BRCA1$req[[1]])
+myseq
+dotPlot( myseq, myseq)
+length( myseq)
+BRCA.variant1 <- getSequence(BRCA1$req[[1]])
+length( BRCA.variant1)
+query(listname = "BRCA1", query="SP=homo sapiens AND K=BRCA1")
+BRCA.variant2 <- getSequence(BRCA1$req[[1]])
+setwd("~/Google Drive/MSc MSE/Visualization and data modelling/GitHubCode/Lecture 2")
+# Exercise 1: download the example2.txt file
+# disclaimer: the data should be in the current directory to run this file
+# Exercise 2: Read this data into R
+data = read.table( "example2.txt", header=TRUE)
+# Exercise 3: Print out the data for cases 10 to 18
+data[10:18, ]
+data
+data[, 2]
+data[c(23,2,5), 2]
+summary( data)
+tapply( data, mean)
+tapply( data$1, mean)
+tapply( data$yrs, mean)
+tapply( data[,1], mean)
+lapply( data, mean)
+lapply( data, c(mean, sum)
+)
+lapply( data, mean)
+sapply( data, mean)
+sapply( data, min)
+sapply( data, max)
+sapply( data, c(max,min,mean)
+)
+sapply( data, sd)
+customizedFunction <- function(x) {
+c(min = min(x), mean = mean(x), max = max(x), sd = sd(x))
+}
+sapply( data, customizedFunction)
+c(min = min(x), mean = mean(x), max = max(x), Deviation = sd(x))
+}
+sapply( data, customizedFunction)
+customizedFunction <- function(x) {
+c(min = min(x), mean = mean(x), max = max(x), Deviation = sd(x))
+}
+sapply( data, customizedFunction)
+source('~/.active-rstudio-document', echo=TRUE)
+random.numbers <- sample( runif( 100, min=0, max=2), size=100, replace=F)
+hist ( random.numbers )
+# Exercise 2: Use dt to evaluate the density function of the t distribution with 13 degrees
+#             of freedom at 20 values in the range -1 to 1.
+density <- dt( x=seq(-1,1,len=20), df=13)
+plot( density )
+# Exercise 3: Find P[X <= x] = 0.01 for a t distribution with 9 degrees of freedom.
+# pt calculates the cumulative distribution function, P( X <= x)
+# the function X is the t distribution with 9 degrees of freedom.
+# x = 0.01
+pt( 0.01, df=9)
+# Exercise 4: IQ scores are known to have a normal distribution with mean 100 and
+#             standard deviation 15. What IQ would you have if you were in the 80th
+min( IQ[ IQ > (0.9 * max(IQ))])
+IQ <- rnorm(1000, mean=100, sd=15)
+quantile( IQ, 0.8) # We apply the quantile function to compute the percentiles
+# we can check graphically the result:
+plot( IQ )
+abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+# Exercise 5: What IQ would you have if you were in the top 10 percent?
+min( IQ[ IQ > (0.9 * max(IQ))])
+IQ
+max(IQ)
+density <- dt( x=seq(-1,1,len=20), df=13)
+plot( density )
+pnorm( 0.9)
+pnorm( 0.9:1.0)
+pnorm( 1.0)
+x = seq( -1, 1, length=100)
+x
+hx = dnorm( x )
+plot( hx, x)
+plot( x, hx)
+px = pnorm( x )
+plot( x, px)
+hx = dnorm( x ); plot( x, hx)
+px = pnorm( x ); plot( x, px)
+x <- seq(-4,4,length=100)*sd + mean
+hx <- dnorm(x,mean,sd)
+IQ <- dnorm(1000, mean=100, sd=15)
+quantile( IQ, 0.8) # We apply the quantile function to compute the percentiles
+IQ <- rnorm(1000, mean=100, sd=15)
+quantile( IQ, 0.8) # We apply the quantile function to compute the percentiles
+mean <- 100
+sd <- 15
+IQ <- seq( -1, 1, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+mean <- 100
+sd <- 15
+IQ <- seq( -1, 1, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+#             percentile?
+mean <- 100
+sd <- 15
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+quantile( dIQ, 0.8)
+quantile( IQ, 0.8)
+quantile( IQ, 0.8, lty="longdash")
+plot( IQ, dIQ, lty="longdash")
+plot( IQ, dIQ)
+quantile( IQ, 0.8)
+# we can check graphically the result:
+plot( IQ )
+abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+plot( IQ )
+abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+abline( a=0, b=80, col="red", lwd=2, lty="longdash")
+plot( IQ )
+#abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+points( x=80, y=(quantile(IQ, 0.8)))
+points( x=80, y=(quantile(IQ, 0.8)), pch=23)
+plot( IQ )
+#abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+points( x=80, y=(quantile(IQ, 0.8)), pch=23)
+plot( IQ, dIQ)
+quantile( IQ, 0.8)
+# we can check graphically the result:
+plot( IQ )
+#abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+points( x=80, y=(quantile(IQ, 0.8)), pch=23)
+plot( IQ, type='l' )
+#abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+points( x=80, y=(quantile(IQ, 0.8)), pch=23)
+plot( IQ, type='l', lty="longdash")
+#abline( a=quantile(IQ, 0.8), b=0, col="red", lwd=2, lty="longdash")
+points( x=80, y=(quantile(IQ, 0.8)), pch=23)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+text( x=80, y=result, "El resultado")
+result <- quantile( IQ, 0.8)
+# we can check graphically the result:
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+text( x=80, y=result, "El resultado")
+text( x=80, y=result, "El resultado", offset=0.5)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+text( x=80, y=result, "El resultado", offset=0.5)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = 'hola'
+text( x=80, y=result, string, offset=0.5)
+string = 'Hola ' + 20
+string = 'Hola ' + as.string(20)
+string = 'Hola %d', 10
+string = 'Hola ' + ' adios'
+string = c('Hola ',' adios')
+text( x=80, y=result, string, offset=0.5)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = c('Hola ',' adios')
+text( x=80, y=result, string, offset=0.5)
+string = c('Hola ',' adios')
+string
+string = as.character( c('Hola ',' adios'))
+string
+string = paste('Hola ',' adios')
+string
+string = paste('Hola ','adios')
+string
+string = paste('Hola ',20)
+string
+string = paste( 80, result)
+string
+string = paste( '(',80,',', result,')')
+string
+string = paste( '(', 80 ,',', result,')')
+string
+text( x=80, y=result, string, offset=0.5)
+text( x=80, y=result, string, offset=0.5, cex=0.6)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = paste( '(', 80 ,',', result,')')
+string
+text( x=80, y=result, string, offset=0.5, cex=0.6)
+text( x=85, y=result, string, offset=0.5, cex=0.6)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = paste( '(', 80 ,',', result,')')
+string
+text( x=85, y=result, string, offset=0.5, cex=0.6)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = paste( '(', 80 ,',', result,')')
+string
+text( x=88, y=result, string, offset=0.5, cex=0.6)
+plot( IQ, type='l', lty="longdash")
+points( x=80, y=result, pch=23)
+string = paste( '(', 80 ,',', result,')')
+string
+text( x=88, y=result, string, offset=0.5, cex=0.6)
+min(IQ)
+max(IQ)
+0.9*max(IQ)
+0.9*(max(IQ) - min(IQ))
+mean <- 100
+sd <- 15
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+0.9*(max(IQ) - min(IQ)) + min(IQ)
+p <- 0.9*(max(IQ) - min(IQ)) + min(IQ)
+min( IQ[ IQ > p])
+p <- 0.9*(max(IQ) - min(IQ)) + min(IQ)
+sprintf("Answer %.2f", p)
+answer <- pt( 0.01, df=9)
+sprintf("Answer %.2f", answer)
+pnorm(142, mean=mean, sd=sd)
+1 - pnorm(142, mean=mean, sd=sd)
+sprintf("Answer %.2f", answer)
+answer <- (1 - pnorm(142, mean=mean, sd=sd))
+sprintf("Answer %.2f", answer)
+sprintf("Answer %.6f", answer)
+sprintf("Answer %.6f %", answer)
+sprintf("Answer %.6f '%'", answer)
+set.seed(144)
+values = runif( 100, min=0, max=2)
+random.numbers <- sample( values, size=20, replace=F)
+set.seed(144)
+values = runif( 100, min=0, max=2)
+set1 <- sample( values, size=20, replace=F)
+set2 <- sample( values, size=20, replace=F)
+set1
+set2
+set.seed(0)
+set1 <- rnorm(n=10, mean = 100, sd=15)
+set2 <- rnorm(n=10, mean = 100, sd=15)
+set1
+set2
+set.seed(0)
+set1 <- rnorm(n=20, mean = 100, sd=15)
+set2 <- rnorm(n=20, mean = 100, sd=15)
+set1
+set2
+set.seed(48639158)
+set1 <- rnorm(n=20, mean = 100, sd=15)
+set2 <- rnorm(n=20, mean = 100, sd=15)
+set1
+set2
+set.seed(0)
+set1 <- rnorm(n=20, mean = 100, sd=15)
+set2 <- rnorm(n=20, mean = 100, sd=15)
+sprintf("First pair of samples")
+set1
+set2
+set.seed(48639158)
+set1 <- rnorm(n=20, mean = 100, sd=15)
+set2 <- rnorm(n=20, mean = 100, sd=15)
+sprintf("Second pair of samples")
+set1
+set2
+mean <- 100
+sd <- 15
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+result <- quantile( IQ, 0.8)
+sprintf("Answer %.2f", result)
+IQ <- seq( -4, 4, length=100)
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+result <- quantile( IQ, 0.8)
+mean <- 100
+sd <- 15
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+result <- quantile( IQ, 0.8)
+mean <- 100
+sd <- 15
+IQ <- seq( -4, 4, length=100)*sd + mean
+dIQ <- dnorm(IQ, mean = mean, sd = sd)
+plot( IQ, dIQ)
+result <- quantile( IQ, 0.8)
+sprintf("Answer %.2f", result)
+foo <- function(x, ...){}
+foo <- function(x, ...){}
+foo <- function(x, ...){
+}
+foo <- function(x, ...)
+{
+ans <- rnorm(x, ...)
+}
+x = 10
+plot( x, foo(x))
+plot(foo(x))
+plot(foo(x, 100, 15))