JMPIN Instruction for Chapter 18
- Example 18.1(pp.684)
Open file XM18-01
Click Analyze and Fit Model, make MARGIN be the Y and Add ROOMS, NEAREST, OFFICE, COLLEGE, INCOME, DISTWIN into the Model Effects Box, then click Run Model.
To see the confidence and prediction intervals, add one more row, then input the value of 3815, 3.4, 476, 24.5, 39, 3.6 for ROOMS, NEAREST, OFFICE, COLLEGE, INCOME, DISTWIN correspondingly. Repeat step 2 above, click on the red triangle in front of Response MARGIN, select Save Columns and pick Mean Confidence Interval and Indiv Confidence Interval (prediction interval), you will have 95% confidence interval as (32.97,41.21) and prediction interval as (25.40, 48.79).
Response MARGIN
Whole Model
Actual by Predicted Plot
ROOMS
Leverage Plot
NEAREST
Leverage Plot
OFFICE
Leverage Plot
COLLEGE
Leverage Plot
INCOME
Leverage Plot
DISTTWN
Leverage Plot
Whole Model
Summary of Fit
RSquare / 0.525062RSquare Adj / 0.49442
Root Mean Square Error / 5.512084
Mean of Response / 45.739
Observations (or Sum Wgts) / 100
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 6 / 3123.8320 / 520.639 / 17.1358
Error / 93 / 2825.6259 / 30.383 / Prob > F
C. Total / 99 / 5949.4579 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 72.454611 / 7.893104 / 9.18 / <.0001
ROOMS / -0.007618 / 0.001255 / -6.07 / <.0001
NEAREST / -1.646237 / 0.632837 / -2.60 / 0.0108
OFFICE / 0.0197655 / 0.00341 / 5.80 / <.0001
COLLEGE / 0.2117829 / 0.133428 / 1.59 / 0.1159
INCOME / -0.413122 / 0.139552 / -2.96 / 0.0039
DISTTWN / 0.2252581 / 0.178709 / 1.26 / 0.2107
- Example 18.2(pp700)
Open file XM18-02
Click Analyze and Fit Models; make Price be Y and Add Bedrooms and H Size and Lot Size into the box of model effects, then click Run Model.
Whole Model
Actual by Predicted Plot
Bedrooms
Leverage Plot
H Size
Leverage Plot
Lot Size
Leverage Plot
Response Price
Whole Model
Summary of Fit
RSquare / 0.559998RSquare Adj / 0.546248
Root Mean Square Error / 25022.71
Mean of Response / 154066
Observations (or Sum Wgts) / 100
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 3 / 7.65017e10 / 2.5501e10 / 40.7269
Error / 96 / 6.0109e+10 / 626135896 / Prob > F
C. Total / 99 / 1.36611e11 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 37717.595 / 14176.74 / 2.66 / 0.0091
Bedrooms / 2306.0808 / 6994.192 / 0.33 / 0.7423
H Size / 74.296806 / 52.97858 / 1.40 / 0.1640
Lot Size / -4.363783 / 17.024 / -0.26 / 0.7982
- Example 18.3(pp705)
Open file XM18-03
Click Analyze and Fit Y by X; make Mark by Y and Time be X, then click OK.
Click Bivariate Fit of Mark by Time and fit line.
To check for the normality of residual, first click on the red triangle in front of Linear Fit and select Save Residuals, then go back to the original data table and you will find there is one more column named Residuals Mark, Click Analyze and Distribution, then put in Residuals Mark as Y, click OK, it appears the histogram of the residuals, click on the red triangle in front of Residuals Mark and select Normal Quantile Plot.
Note: to change Mark to log (Mark), click Bivariate Fit of Mark by Time and Fit Special, then select Y Transformation: Natural Logarithm, click OK.
Note: to change Mark to 1/Mark, click Bivariate Fit of Mark by Time and Fit Special, then select Y Transformation: Reciprocal, click OK.
Bivariate Fit of Mark By Time
Linear Fit
Mark = -2.2 + 0.55 Time
Summary of Fit
Rsquare / 0.743974RSquare Adj / 0.741362
Root Mean Square Error / 2.304609
Mean of Response / 25.3
Observations (or Sum Wgts) / 100
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 1 / 1512.5000 / 1512.50 / 284.7743
Error / 98 / 520.5000 / 5.31 / Prob > F
C. Total / 99 / 2033.0000 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / -2.2 / 1.64582 / -1.34 / 0.1844
Time / 0.55 / 0.032592 / 16.88 / <.0001
Distributions
Residuals Mark
Quantiles
100.0% / maximum / 8.200099.5% / 8.2000
97.5% / 6.1500
90.0% / 2.9250
75.0% / quartile / 1.4500
50.0% / median / -0.0500
25.0% / quartile / -1.5500
10.0% / -2.7750
2.5% / -4.1688
0.5% / -5.8000
0.0% / minimum / -5.8000
Moments
Mean / 7.105e-16Std Dev / 2.2929404
Std Err Mean / 0.229294
upper 95% Mean / 0.4549718
lower 95% Mean / -0.454972
N / 100
Transformed Fit Log
Log(Mark) = 2.1295821 + 0.0217159 Time
Summary of Fit
RSquare / 0.771412RSquare Adj / 0.769079
Root Mean Square Error / 0.084437
Mean of Response / 3.215377
Observations (or Sum Wgts) / 100
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 1 / 2.3579010 / 2.35790 / 330.7181
Error / 98 / 0.6987048 / 0.00713 / Prob > F
C. Total / 99 / 3.0566058 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 2.1295821 / 0.0603 / 35.32 / <.0001
Time / 0.0217159 / 0.001194 / 18.19 / <.0001
Transformed Fit Recip
Recip(Mark) = 0.0845743 - 0.0008765 Time
Summary of Fit
RSquare / 0.777884RSquare Adj / 0.775617
Root Mean Square Error / 0.003345
Mean of Response / 0.040751
Observations (or Sum Wgts) / 100
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 1 / 0.00384099 / 0.003841 / 343.2104
Error / 98 / 0.00109675 / 0.000011 / Prob > F
C. Total / 99 / 0.00493774 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 0.0845743 / 0.002389 / 35.40 / <.0001
Time / -0.000876 / 0.000047 / -18.53 / <.0001
- Example 18.4(pp717)
Open file XM18-04
Click Analyze and Fit Models; make Tickets be Y and Add Snowfall and Temperature into the box of model effects, then click Run Model.
Click Response Tickets, Row Diagnostics and Durbin Watson Test.
Response Tickets
Whole Model
Actual by Predicted Plot
Snowfall
Leverage Plot
Temperature
Leverage Plot
Response Tickets
Whole Model
Summary of Fit
RSquare / 0.12003RSquare Adj / 0.016504
Root Mean Square Error / 1711.676
Mean of Response / 9315.3
Observations (or Sum Wgts) / 20
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 2 / 6793798 / 3396899 / 1.1594
Error / 17 / 49807214 / 2929836 / Prob > F
C. Total / 19 / 56601012 / 0.3373
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 8308.0114 / 903.7285 / 9.19 / <.0001
Snowfall / 74.593249 / 51.57483 / 1.45 / 0.1663
Temperature / -8.753738 / 19.70436 / -0.44 / 0.6625
Effect Tests
Source / Nparm / DF / Sum of Squares / F Ratio / Prob > FSnowfall / 1 / 1 / 6128677.7 / 2.0918 / 0.1663
Temperature / 1 / 1 / 578236.9 / 0.1974 / 0.6625
Durbin-Watson
Durbin-Watson / Number of Obs. / AutoCorrelation0.5931403 / 20 / 0.5914
Add new column and fill in the corresponding cell with 1, 2, …, 20. And change the name of the column to “Years”. Then Click Analyze and Fit Models; make Tickets be Y and Add Snowfall, Temperature and Years into the box of model effects, then click Run Model.
Response Tickets
Whole Model
Actual by Predicted Plot
Summary of Fit
RSquare / 0.74098RSquare Adj / 0.692414
Root Mean Square Error / 957.2354
Mean of Response / 9315.3
Observations (or Sum Wgts) / 20
Analysis of Variance
Source / DF / Sum of Squares / Mean Square / F RatioModel / 3 / 41940217 / 13980072 / 15.2571
Error / 16 / 14660795 / 916299.68 / Prob > F
C. Total / 19 / 56601012 / <.0001
Parameter Estimates
Term / Estimate / Std Error / t Ratio / Prob>|t|Intercept / 5965.5876 / 631.2518 / 9.45 / <.0001
Snowfall / 70.183059 / 28.85142 / 2.43 / 0.0271
Temperature / -9.232802 / 11.01971 / -0.84 / 0.4145
Years / 229.96997 / 37.13209 / 6.19 / <.0001
Effect Tests
Source / Nparm / DF / Sum of Squares / F Ratio / Prob > FSnowfall / 1 / 1 / 5422102 / 5.9174 / 0.0271
Temperature / 1 / 1 / 643227 / 0.7020 / 0.4145
Years / 1 / 1 / 35146419 / 38.3569 / <.0001
Residual by Predicted Plot
Snowfall
Leverage Plot
Temperature
Leverage Plot
Years
Leverage Plot