This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Statistics 224/324  lecture 1 (Bates) Midterm 2 20080410 (p. 1) The kfm data set from the text gives the milk intake (dl/day) of 50 infants approximately 2 months old; 25 girls and 25 boys. boy girl 5 6 7 8 9 10 ● ● Milk intake (dl/day) Density 0.0 0.1 0.2 0.3 4 6 8 10 12 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● boy girl ● Standard normal quantiles Milk intake (dl/day) 5 6 7 8 9 1021 1 2 ● ● ● ● ●● ●● ●●● ● ●●●●●● ● ● ● ● ● ● ● > with(kfm, summary(dl.milk)) Min. 1st Qu. Median Mean 3rd Qu. Max. 4.440 6.555 7.660 7.504 8.428 10.430 > with(kfm, c(variance = var(dl.milk), std.dev. = sd(dl.milk))) variance std.dev. 2.284666 1.511511 > t.test(dl.milk ~ sex, kfm, conf.level = 0.99) Welch Two Sample ttest data: dl.milk by sex t = 2.174, df = 47.89, pvalue = 0.03468 alternative hypothesis: true difference in means is not equal to 0 99 percent confidence interval:0.2095362 2.0015362 sample estimates: mean in group boy mean in group girl 7.9524 7.0564 The article “Pavement thickness designs for nofines concrete parking lots” ( J. of Transportation Engr. , 1995: 476484) provided data from an experiment measuring y = porosity (%) as it relates to x = unit weight (pcf) in concrete specimens. Unit weight (pcf) Porosity (%) 10 15 20 25 100 105 110 115 120 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Fitted values Residuals1 1 10 15 20 25 30 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Standard normal quantiles Residuals1 121 1 2 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Statistics 224/324  lecture 1 (Bates) Midterm 2 20080410 (p. 2) > summary(fm1 < lm(porosity ~ weight, concrete)) Coefficients: Estimate Std. Error t value Pr(>t) (Intercept) 118.90992 4.49912 26.43 1.10e12 weight0.90473 0.0410922.02 1.12e11 Residual standard error: 0.938 on 13 degrees of freedom Multiple Rsquared: 0.9739, Adjusted Rsquared: 0.9719 Fstatistic: 484.8 on 1 and 13 DF, pvalue: 1.125e11 > confint(fm1) 2.5 % 97.5 % (Intercept) 109.1901530 128.6296806 weight0.99349660.8159647 > anova(fm1) Df Sum Sq Mean Sq F value Pr(>F) weight 1 426.62 426.62 484.84 1.125e11 Residuals 13 11.44 0.88 > unlist(predict(fm1, list(weight = 115), se = TRUE)) fit.1 se.fit df residual.scale 14.8658911 0.3357749 13.0000000 0.9380352 > qt(c(0.05, 0.025, 0.01, 0.005), df = 13, lower = FALSE) [1] 1.770933 2.160369 2.650309 3.012276 A consumer producttesting organization wishes to compare the annual power consumption for different brands of dehumidifier. Because power consumption depends on the prevailing humidity level, it was decided to monitor each brand at four different levels of humidity ranging from mod erate to very high humidity. Each brand was tested at each humidity level and the annual power consumption (kWh) recorded....
View
Full
Document
This note was uploaded on 03/06/2010 for the course STATS 324 taught by Professor Bates during the Fall '08 term at University of Wisconsin.
 Fall '08
 Bates
 Statistics

Click to edit the document details