chapter 6
DESCRIPTION
Chapter 6. Ratio, Regression and Difference Estimation. 6.2 Ratio Estimators. Looking at two measurements of a population, call these x and y (we will only focus on the SRS sampling strategy) Note that the ratio of population means is the same as the ratio of population totals!! - PowerPoint PPT PresentationTRANSCRIPT
Chapter 6
Ratio, Regression and Difference Estimation
6.2 Ratio Estimators
• Looking at two measurements of a population, call these x and y (we will only focus on the SRS sampling strategy)
• Note that the ratio of population means is the same as the ratio of population totals!!– x/y = Nx/Ny = x/y
• Therefore, we will use R = ratio of the means
6.3 Ratio estimator for SRS
• Calculating an estimate and bound on r:– r = ybar/xbar
• Estimated variance of r:– (1-n/N)*(1/x
2)*(sr2/n)
– where sr2 = i(yi-rxi)2/(n-1)
– Use xbar to estimate x when necessary
Ratio estimator of total and mean
• For population total:– haty = rx
• Estimated variance of haty
– N2(1-n/N) sr2 /n (where sr
2 is defined previously)
• For population mean:– haty = rx
• Estimated variance of haty
– (1-n/N) sr2 /n (where sr
2 is defined previously)
Examples
• 6.1: mean of x (square-foot) = 0.558333 mean of y ( volume) = 11.83333haty = (11.8333/.558333)*75=1589.552204sr
2 = 1.750299Estimated variance = 250^2*(1-12/250)*
1.750299/12 = 8678.566Bound=2*sqrt(8678.566) = 186.3176Without fpc (don’t really need it since 1-n/N =
0.952), B = 190.957
More examplesRcode:x=dat$Basal.ary=dat$Volumeybar=mean(y)xbar = mean(x)rhat =(ybar/xbar)t.hat.y = rhat*75n=length(dat$Volume)s.r.2 = sum((y-rhat*x)^2)/(n-1)var.that = (250^2)*(1-12/250)*s.r.2/12___________________________________________________• 6.2: that =250* 11.83333 = 2958.333• Variance for that=250^2*(1-12/250)*26.87879/12 = 133274• Bound = 730.1342• Without fpc, bound = 748.3146
More examples
• Do problem 6.6 mean(x)=16.47273,mean(y) = 16.84545, so
rhat = 16.84545/16.47273=1.022627; sr
2=0.2049424
• Coefficient of variation: CV(x) = sx/xbar, where sx is the standard deviation of x. It is desired that the CV is lower than 1. Find the CV for x and the CV for y in problem 6.6.
6.4 Sample size estimation
• n = N2/(ND + 2)where we can estimate 2 as
i(yi-rxi)2/(n’-1) with n’ a small preliminary sample
With D=B2x2/4 for estimating R
D = B2/4 for estimating y
D = B2/(4N2) for estimating y
Do problem 6.13