Automobile Data Analysis and Visualization

The script begins by creating a temporary SAS^© dataset named 'auto' using a DATA statement and in-line data via CARDS. This dataset contains vehicle information such as make, MPG, reliability (rep78), weight, and origin (foreign1). It then uses PROC FREQ to obtain the frequency distribution of the 'mpg' variable. After that, it initializes global graphic options. Two types of graphs are generated: a simple PROC PLOT to visualize 'mpg' as a function of 'weight1', and a more advanced PROC GPLOT. The customized PROC GPLOT displays the same relationship but adds a visual distinction based on the 'foreign1' variable (foreign or non-foreign vehicle), defines specific axes, and calculates and displays the regression equation (regeqn) for the fitted line.

Data Analysis

Type : CREATION_INTERNE

The data is directly integrated into the script via a DATA statement with CARDS, creating the temporary 'auto' dataset.

1 Code Block

DATA STEP Data

Explanation :
This DATA STEP block creates a temporary SAS dataset named 'auto'. Data is read in-line using the CARDS statement and defines the variables 'make' (character), 'mpg', 'rep78', 'weight1', and 'foreign1' (numeric). 'mpg' represents fuel consumption, 'weight1' is weight, and 'foreign1' indicates whether the car is foreign (1) or not (0).

Copied!

1	DATA auto ;
2	INPUT make $ mpg rep78 weight1 foreign1 ;
3	CARDS;
4	AMC 22 3 2930 0
5	AMC 17 3 3350 0
6	AMC 22 . 2640 0
7	Audi 17 5 2830 1
8	Audi 23 3 2070 1
9	BMW 25 4 2650 1
10	Buick 20 3 3250 0
11	Buick 15 4 4080 0
12	Buick 18 3 3670 0
13	Buick 26 . 2230 0
14	Buick 20 3 3280 0
15	Buick 16 3 3880 0
16	Buick 19 3 3400 0
17	Cad. 14 3 4330 0
18	Cad. 14 2 3900 0
19	Cad. 21 3 4290 0
20	Chev. 29 3 2110 0
21	Chev. 16 4 3690 0
22	Chev. 22 3 3180 0
23	Chev. 22 2 3220 0
24	Chev. 24 2 2750 0
25	Chev. 19 3 3430 0
26	Datsun 23 4 2370 1
27	Datsun 35 5 2020 1
28	Datsun 24 4 2280 1
29	Datsun 21 4 2750 1
30	;
31	RUN;

2 Code Block

PROC FREQ

Explanation :
This procedure generates a frequency table for the 'mpg' variable from the 'auto' dataset. It allows observing the distribution of different fuel consumption values.

Copied!

1	PROC FREQ DATA = auto;
2	TABLES mpg;
3	RUN;

3 Code Block

GOPTIONS

Explanation :
This statement resets all global graphic options to their default values and adds a border to the generated graphs. This ensures a clean base for subsequent graphs.

Copied!

1	goptions reset=all border;

4 Code Block

PROC PLOT

Explanation :
This procedure generates a simple scatter plot ('plot') of 'mpg' (Y-axis) versus 'weight1' (X-axis) from the 'auto' dataset. It provides an initial visual overview of the relationship between these two variables.

Copied!

1
2	PROC PLOT
3	DATA=auto;
4	plot mpg * weight1 ;
5
6	RUN;
7

5 Code Block

PROC GPLOT

Explanation :
This procedure generates a more elaborate GPLOT graph. The title 'Study of MPG vs Weight' is set. The SYMBOL statement configures the display of points and the regression line (interpol=rqcli95 for a quadratic regression with 95% confidence intervals, points as circles, specific colors). The graph represents 'mpg' as a function of 'weight1', with points colored differently according to the 'foreign1' variable. The X and Y axes are customized with specific ranges and increments, and the 'regeqn' option displays the regression equation on the graph.

Copied!

1	PROC GPLOT DATA=auto;
2	title "Study of MPG vs Weight";
3
4	symbol interpol= rqcli95
5	value=circle
6	cv= crimson
7	ci = black
8	co = bib
9	width= 2
10	;
11
12	plot mpg*weight1 = foreign1 / haxis=2000 to 4500 BY 500
13	vaxis=12 to 35 BY 2
14	regeqn;
15
16	RUN;

This material is provided "as is" by We Are Cas. There are no warranties, expressed or implied, as to merchantability or fitness for a particular purpose regarding the materials or code contained herein. We Are Cas is not responsible for errors in this material as it now exists or will exist, nor does We Are Cas provide technical support for it.

Copyright Info : Author - Anupama Rajaram Program Description - This program creates a simple gplot of 2 variables, draws the plot line and calculates regression equation. y-axis = mpg. x-axis = weight1.

Retour à la liste