Question

An article in Tec namet ics b S С. Narula and·F. Welington Prediction, Linea Regress on, and Minimum Sum o Ralatva Erro vol. 19 1977 prasanti data on the salling price y and annual taxas for 24 hous" The taxas indude cca acho and county taxes. Tha data are aho n in tab a o e n Sele Pric 1000 T/10co 29-5 5.0208 29.9 5.0597 30.9 5.898 1.5 5.3003 30.9 5.9592 29.6 4.8420 6.6969 37.5 .9894 8.7951 36.9 9.14 Round your ansers to 3 decimal places. Find 99.900% confidence merval on each of the following; Intarcapt. b) Slope. (s) Mean selling price when the taces peidr xo -6.7 (d) Compute the 99.800%á prediction interval fer salling prica when the tasas paid are so-6.7

Answer 1

Regression using Minitab:

Click Stat > Regression > Regression... on the top menu, as shown below:

all File Edit Data Calc Stat Graph Editor ools Window Help Assistant 묘刍XEg Basic Statistics Regressiorn ANOVA DOE Control Charts Regression... General Regression Stepwise... Best Subsets 07101/2 Quality Tools ▶ lut Fitted Line Plot Rejiability/SurvivalNonlingar Regression.. Multivariate Time Series Tables Nonparametrics EDA Power and Sample Size、LN Nominal Logistic Regression Welcome to Minitab, Orthogonal Regressio.. 荐目Partial Least Squares 、 LG Binary Logistic Regression I-a Ordinal Logistic Regression ...

You will be presented with the following Regression dialogue box:

Minitab-Untitled -Worksheet 2** File Edit Data Calc Stat Graph Editor Tools Window Help Assistant .E3 비 r lo t 摊的. S @国 TOOL C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 Sales price/1000 Taxes per 1000 50208 45429 45573 50597 3891 5898 56039 58262 53003 62712 59592 48420 82464 66969 77841 90384 59894 75422 87951 60831 Regression 29.5 27.9 25.9 29.9 29.9 30.9 28.9 35.9 31.5 31.0 30.9 39.6 36.9 41.9 40.5 43.9 37.5 37.9 44.5 37.9 C1 Sales price/1000 Responses: C2 Taxes per 1000 Taxes per Continuous predictors Sales price/1000 10 12 13 15 16 17 18 19 20 21 Options.. Stepwise... Storage.. 0KCancel

Click the OK button. The output that Minitab produces is shown below.

Regression Analysis: Taxes per 1000 versus Sales price/1000 Analysis of Variance Seq SS Contribution Adj SS 34.11% 5006656365 5006656365 34.11% 5006656365 5006656365 65.89% 9672507913 439659451 25.21% 3700449986 205580555 40.68% 5972057927 1493014482 Adj MS F-Value P-Value 0.003 0.003 DF Source Regression 11.39 11.39 1 5006656365 Sales price/1000 1 5006656365 22 9672507913 18 3700449986 4 5972057927 23 14679164277 Error 0.14 0.999 Lack-of-Fit Pure Error Total 100. 00% Model Summary R-sq 20968.1 34.11% R-sq (adj) 31.11% PRESS 1.10377E+10 R-s (pred) 24 . 81% Coefficients Term Constant Sales price/1000 Coef SE Coef 99.8% CI T-Value P-Value VIF 0.272 0.003 1.00 -29103 25827 (-119626, 61420) 95, 5005) 1.13 3.37 2455 728 Regression Equation Taxes per 1000-29103 + 2455 Sales price/1000

a.

99.8% CI for intercept is (-119626, 61420)

98% CI for slope ( -95, 5005)

b.

Regression Equation

Taxes per 1000 = -29103 + 2455 Sales price/1000

=-29103 + 2455*6.7

=-12653.9

Prediction interval:

Prediction for Taxes per 1000 Regression Equation Taxes per 1000 = -29103 + 2455 sales price/ 1000 Variable Sales price/1000 Setting 6.7 Fit SE Fit 99.8% CI 99.8% PI 12653.9 21035.3 (-86382.4, 61074.5) (-116755, 91447.4) xx xx denotes an extremely unusual point relative to predictor levels used to fit the model.