Presentation is loading. Please wait.

Presentation is loading. Please wait.

Interview Example: nknw964.sas Y = rating of a job applicant Factor A represents 5 different personnel officers (interviewers) n = 4.

Published byDarleen Richardson Modified over 8 years ago

Similar presentations

Presentation on theme: "Interview Example: nknw964.sas Y = rating of a job applicant Factor A represents 5 different personnel officers (interviewers) n = 4."— Presentation transcript:

1 Interview Example: nknw964.sas Y = rating of a job applicant Factor A represents 5 different personnel officers (interviewers) n = 4

2 Interview Example: Input data interview; infile ‘H:\My Documents\Stat 512\CH25TA01.DAT'; input rating officer; proc print data=interview; run; Obsratingofficer 1761 2651 3851 4741 5592 6752 7812 8672 9493 10633 11613 12463

3 Interview Example: Scatterplot axis2 label=(angle=90); symbol1 v=circle i=none c=black; proc gplot data=interview; plot rating*officer/vaxis=axis2; run; title1 h=3 'Scatterplot of rating vs. officer'; goptions htext=2;

4 Interview Example: Means plot proc means data=interview; output out=a2 mean=avrate; var rating; by officer; run; title1 h=3 'Plot of the means of rating vs. officer'; symbol1 v=circle i=join c=black; proc gplot data=a2; plot avrate*officer/vaxis=axis2; run;

5 Interview Example: Means plot (cont)

6 Interview Example: random ANOVA proc glm data=interview; class officer; model rating=officer; random officer; run;

7 Interview Example: random ANOVA (cont) SourceDFSum of SquaresMean SquareF ValuePr > F Model41579.700000394.9250005.390.0068 Error151099.25000073.283333 Corrected Total192678.950000 R-SquareCoeff VarRoot MSErating Mean 0.58967111.981208.56056971.45000 SourceDFType III SSMean SquareF ValuePr > F officer41579.700000394.9250005.390.0068 SourceType III Expected Mean Square officerVar(Error) + 4 Var(officer)

8 Interview Example: Variances proc varcomp data=interview; class officer; model rating=officer; run; MIVQUE(0) Estimates Variance Componentrating Var(officer)80.41042 Var(Error)73.28333

9 Interview Example: ANOVA random (mixed) proc mixed data=interview cl; class officer; model rating=; random officer/vcorr; run;

10 Interview Example: ANOVA random (mixed) (cont) The Mixed Procedure Covariance Parameter Estimates Cov ParmEstimateAlphaLowerUpper officer80.41040.0524.45721498.97 Residual73.28330.0539.9896175.54 Estimated V Correlation Matrix for Subject 1 RowCol1Col2Col3Col4 11.00000.5232 2 1.00000.5232 3 1.00000.5232 4 1.0000

11 Efficiency Example: nknw976.sas Y = fuel efficiency in mpg Factor A = 4 drivers Factor B = 5 cars n = 4

12 Efficiency Example: Input goptions htext=2; data efficiency; infile ‘H:\My Documents\Stat 512\CH25PR15.DAT'; input mpg driver car; proc print data=efficiency; run; Obsmpgdrivercar 125.311 225.211 328.912 430.012 524.813 625.113 728.414 827.914 927.115 1026.615

13 Efficiency Example: Scatterplot data efficiency; set efficiency; dc = driver*10 + car; title1 h=3 'Scatterplot'; axis2 label=(angle=90); symbol1 v=circle i=none c=blue; proc gplot data=efficiency; plot mpg*dc/vaxis=axis2; run;

14 Efficiency Example: Scatterplot (cont)

15 Efficiency Example: Interaction Plot proc means data=efficiency; output out=effout mean=avmpg; var mpg; by driver car; title1 h=3 'Interaction Plot'; symbol1 v='A' i=join c=black h=1.5; symbol2 v='B' i=join c=red h=1.5; symbol3 v='C' i=join c=green h=1.5; symbol4 v='D' i=join c=blue h=1.5; symbol5 v='E' i=join c=orange h=1.5; proc gplot data=effout; plot avmpg*driver=car/vaxis=axis2; run;

16 Efficiency Example: Interaction Plot (cont)

17 Efficiency Example: ANOVA proc glm data=efficiency; class driver car; model mpg=driver car driver*car; random driver car driver*car/test; run; SourceDFSum of SquaresMean SquareF ValuePr > F Model19377.444750019.8655132113.03<.0001 Error203.51500000.1757500 Corrected Total39380.9597500 SourceDFType III SSMean SquareF ValuePr > F driver3280.284750093.4282500531.60<.0001 car494.713500023.6783750134.73<.0001 driver*car122.44650000.20387501.160.3715

18 Efficiency Example: ANOVA (cont) SourceType III Expected Mean Square driverVar(Error) + 2 Var(driver*car) + 10 Var(driver) carVar(Error) + 2 Var(driver*car) + 8 Var(car) driver*carVar(Error) + 2 Var(driver*car)

19 Efficiency Example: ANOVA (cont) Tests of Hypotheses for Random Model Analysis of Variance Dependent Variable: mpg SourceDFType III SSMean SquareF ValuePr > F driver3280.28475093.428250458.26<.0001 car494.71350023.678375116.14<.0001 Error122.4465000.203875 Error: MS(driver*car) SourceDFType III SSMean SquareF ValuePr > F driver*car122.4465000.2038751.160.3715 Error: MS(Error)203.5150000.175750

20 Efficiency Example: variances proc varcomp data=efficiency; class driver car; model mpg=driver car driver*car; run; MIVQUE(0) Estimates Variance Componentmpg Var(driver)9.32244 Var(car)2.93431 Var(driver*car)0.01406 Var(Error)0.17575

21 Efficiency Example: ANOVA proc mixed data=efficiency cl; class car driver; model mpg=; random car driver car*driver/vcorr; run; Covariance Parameter Estimates Cov ParmEstimateAlphaLowerUpper car2.93430.051.046424.9038 driver9.32240.052.9864130.79 car*driver0.014060.050.0013453.592E17 Residual0.17570.050.10290.3665

22 Efficiency Example: Interaction Plot (cont)

23 Service Example: 25.16 (nknw1005.sas) Y = service time for disk drives A = make of drive (3) fixed B = technician performing the service (3) random n = 5

24 Service Example: input data service; infile 'H:\My Documents\Stat 512\CH19PR16.DAT'; input time tech make k; mt = make*10+tech; proc print data=service; run; title1 'Proc glm with tech, make*tech random'; proc glm data=service; class make tech; model time = make tech make*tech; random tech make*tech/test; run;

25 Service Example: ANOVA SourceDFSum of SquaresMean SquareF ValuePr > F Model81268.177778158.5222223.050.0101 Error361872.40000052.011111 Corrected Total443140.577778 SourceDFType III SSMean SquareF ValuePr > F make228.31111114.1555560.270.7633 tech224.57777812.2888890.240.7908 make*tech41215.288889303.8222225.840.0010 SourceType III Expected Mean Square makeVar(Error) + 5 Var(make*tech) + Q(make) techVar(Error) + 5 Var(make*tech) + 15 Var(tech) make*techVar(Error) + 5 Var(make*tech)

26 Service Example: /test Tests of Hypotheses for Mixed Model Analysis of Variance Dependent Variable: time SourceDFType III SSMean SquareF ValuePr > F make*tech41215.288889303.8222225.840.0010 Error: MS(Error)361872.40000052.011111 SourceDFType III SSMean SquareF ValuePr > F make228.31111114.1555560.050.9550 tech224.57777812.2888890.040.9607 Error: MS(make*tech)41215.288889303.822222

Download ppt "Interview Example: nknw964.sas Y = rating of a job applicant Factor A represents 5 different personnel officers (interviewers) n = 4."

Similar presentations

Randomized Complete Block and Repeated Measures (Each Subject Receives Each Treatment) Designs KNNL – Chapters 21,27.1-2.

Randomized Complete Block and Repeated Measures (Each Subject Receives Each Treatment) Designs KNNL – Chapters 21,
Topic 32: Two-Way Mixed Effects Model. Outline Two-way mixed models Three-way mixed models.

Topic 32: Two-Way Mixed Effects Model. Outline Two-way mixed models Three-way mixed models.
Statistical Analysis Professor Lynne Stokes Department of Statistical Science Lecture #19 Analysis of Designs with Random Factor Levels.

Statistical Analysis Professor Lynne Stokes Department of Statistical Science Lecture #19 Analysis of Designs with Random Factor Levels.
Weekend Workshop I PROC MIXED. Random or Fixed ?RANDOMFIXEDLevels: Selected at random from infinite population Finite number of possibilities Another.

Weekend Workshop I PROC MIXED. Random or Fixed ?RANDOMFIXEDLevels: Selected at random from infinite population Finite number of possibilities Another.
CS Example: General Linear Test (cs2.sas)

CS Example: General Linear Test (cs2.sas)
Topic 3: Simple Linear Regression. Outline Simple linear regression model –Model parameters –Distribution of error terms Estimation of regression parameters.

Topic 3: Simple Linear Regression. Outline Simple linear regression model –Model parameters –Distribution of error terms Estimation of regression parameters.
Comparing Regression Lines

Comparing Regression Lines
Creating Graphs on Saturn GOPTIONS DEVICE = png HTITLE=2 HTEXT=1.5 GSFMODE = replace; PROC REG DATA=agebp; MODEL sbp = age; PLOT sbp*age; RUN; This will.

Creating Graphs on Saturn GOPTIONS DEVICE = png HTITLE=2 HTEXT=1.5 GSFMODE = replace; PROC REG DATA=agebp; MODEL sbp = age; PLOT sbp*age; RUN; This will.
Partial Regression Plots. Life Insurance Example: (nknw364.sas) Y = the amount of life insurance for the 18 managers (in $1000) X 1 = average annual income.

Partial Regression Plots. Life Insurance Example: (nknw364.sas) Y = the amount of life insurance for the 18 managers (in $1000) X 1 = average annual income.
Some Terms Y =  o +  1 X Regression of Y on X Regress Y on X X called independent variable or predictor variable or covariate or factor Which factors.

Some Terms Y =  o +  1 X Regression of Y on X Regress Y on X X called independent variable or predictor variable or covariate or factor Which factors.
Multiple regression analysis

Multiple regression analysis
Matrix A matrix is a rectangular array of elements arranged in rows and columns Dimension of a matrix is r x c  r = c  square matrix  r = 1  (row)

Matrix A matrix is a rectangular array of elements arranged in rows and columns Dimension of a matrix is r x c  r = c  square matrix  r = 1  (row)
HIERARCHICAL LINEAR MODELS USED WITH NESTED DESIGNS IN EDUCATION, PSYCHOLOGY USES RANDOM FACTORS EXPECTED MEAN SQUARE THEORY COMBINES INFORMATION ACROSS.

HIERARCHICAL LINEAR MODELS USED WITH NESTED DESIGNS IN EDUCATION, PSYCHOLOGY USES RANDOM FACTORS EXPECTED MEAN SQUARE THEORY COMBINES INFORMATION ACROSS.
LECTURE 13 ANALYSIS OF COVARIANCE AND COVARIANCE INTERACTION and ATI (Aptitude-Treatment Interaction)

LECTURE 13 ANALYSIS OF COVARIANCE AND COVARIANCE INTERACTION and ATI (Aptitude-Treatment Interaction)
Purpose to compare average response among levels of the factors Chapters 2-4 -predict future response at specific factor level -recommend best factor.

Purpose to compare average response among levels of the factors Chapters 2-4 -predict future response at specific factor level -recommend best factor.
Example: Price Analysis for Diamond Rings in Singapore Variables Response Variable: Price in Singapore dollars (Y) Explanatory Variable: Weight of diamond.

Example: Price Analysis for Diamond Rings in Singapore Variables Response Variable: Price in Singapore dollars (Y) Explanatory Variable: Weight of diamond.
Topic 18: Model Selection and Diagnostics

Topic 18: Model Selection and Diagnostics
Topic 16: Multicollinearity and Polynomial Regression.

Topic 16: Multicollinearity and Polynomial Regression.
Topic 28: Unequal Replication in Two-Way ANOVA. Outline Two-way ANOVA with unequal numbers of observations in the cells –Data and model –Regression approach.

Topic 28: Unequal Replication in Two-Way ANOVA. Outline Two-way ANOVA with unequal numbers of observations in the cells –Data and model –Regression approach.
1 Experimental Statistics - week 7 Chapter 15: Factorial Models (15.5) Chapter 17: Random Effects Models.

1 Experimental Statistics - week 7 Chapter 15: Factorial Models (15.5) Chapter 17: Random Effects Models.

Similar presentations

Ads by Google