{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# hw8sol - Stat 511 Homework 8 solution Spring 2011 1 The...

This preview shows pages 1–2. Sign up to view the full content.

Stat 511 Homework 8 solution Spring 2011 1. The data in range.txt come from a field study of plant response to fertilization. There are five treatments (a control and four different fertilization regimes) randomly assigned to plots arranged into blocks. The response is the average PO4 concentration in plant leaves. (a) Assume blocks are a fixed effect and analyze the data. The model with fixed block effects reads Y ij = μ + α i + β j + ij with ij N (0 , σ 2 e ), i, j = 1 , · · · 5. Report: The F-statistic and p-value for the test of no differences among fertilizers are reported by R output as: 22 . 605 and 2 × 10 - 6 The estimated difference between between the 100lb.N treatment and the 100lbNP treatments is given by ¯ Y 2 · - ¯ Y 1 · , and by R output it is 3 . 80 with standard error 0 . 5648. The mean for the No fertilizer (No.fert) treatment (averaged over blocks) is 8.06 ( ¯ Y 5 · ) with standard error 0 . 3994. (b) Assume blocks are a random effect and analyze the data. Y ij = μ + α i + β j + ij with ij N (0 , σ 2 e ) , β j N (0 , σ 2 b ), i, j = 1 , · · · 5 Report: The F-statistic and p-value are the same as part(a). So are the estimated difference between between the 100lb.N treatment and the 100lbNP treatments and its standard error. The mean for the No fertilizer (No.fert) treatment (averaged over blocks) is 8.06 with standard error p ( σ 2 e + σ 2 b ) / 5 = 0 . 8678. (c) Using the default parameterizations, R reports an estimated intercept of 4.2520 when using a fixed block analysis and 6.5800 when using a random block analysis. Explain why these numbers are different and how you can extract 6.5800 from the estimates provided by the fixed block analysis. The random block analysis calculates the intercept by averaging over all the blocks and the fixed block analysis only using data from block 1. Thus 6 . 58 = 4 . 252 + 5 X i =2 ˆ β i / 5 = 4 . 252 + (2 . 02 + 1 . 88 + 2 . 86 + 4 . 88 + 0) / 5 2. The data in ryegrass.txt come from a study of pasture management in New Zealand. The

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}