Statistics Computer Lab 1
Internet : Histograms, Regression and Correlation
Name:______________________________ Period:______ Date:____________
Write your answers on a separate sheet of paper for III and IV.
I.
Guessing Correlations – Have your best score out of 20 verified by your teacher. You
may start over as many times as you wish.
_______/20
http://istics.net/stat/Correlations/
II.
Regression - see the effects of adding an outlier.
Write the new LSRL equation and r-value obtained by adding each of the following
points:
a. (10, 200)
LSRL:______________________________ r = ________
b. (100, 200)
LSRL:______________________________ r = ________
c. (200, 200)
LSRL:______________________________ r = ________
d. (200, 10)
LSRL:______________________________ r = ________
e. Which point(s) seem to have the greatest influence on changing the LSRL?
http://www.stat.sc.edu/~west/javahtml/Regression.html
III.
Histogram - check the effect of bin size on Old Faithful data
Describe how the distribution changes when number of bins change from 2 upwards.
http://www.stat.sc.edu/~west/javahtml/Histogram.html
IV.
See how different points cause new equations.
http://mste.illinois.edu/users/carvell/PlotPoints/default.html
Enter your own points to see the regression equation. Do three iterations, and
change your data set each time.
For each iteration, list the data points .
a.
b.
c.
V. http://onlinestatbook.com/stat_sim/
a. Go to the Regression by Eye applet. Try drawing in your own best guess for the
regression line, and see how good you are. Try 5 times to see if you improve.
1. My MSE =__________________ True MSE = ________________
2. My MSE =__________________ True MSE = ________________
3. My MSE =__________________ True MSE = ________________
4. My MSE =__________________ True MSE = ________________
5. My MSE =__________________ True MSE = ________________
b. Go to the Transformations. Try different combinations of transformations on the X and
Y variables. Which ones delivered a better linear regression? Try 3 different data sets:
Data set 1:____________________________
What transformation combination worked best?
Regression Equation:_________________________________ r = ________
Data set 2: __________________________
What transformation combination worked best?
Regression Equation:_________________________________ r = ________
Data set 3: __________________________
What transformation combination worked best?
Regression Equation:_________________________________ r = ________