Correlation & Regression • CR12

Correlation and Regression Numerical Assignment Practice

Past-paper numerical practice for Correlation and Regression. Try each question first, then click Check Answer.

Assignment 1 — Section B Numericals

This page contains only numerical practice questions. Theory questions are kept in CR13.
1

Coding Hours and Practical Score

Find the relationship between coding hours per week and practical score. Also find the regression equation and interpret the result.

Coding Hours per Week \(X\): 2, 5, 7, 10, 13, 17
Practical Score \(Y\): 52, 60, 66, 72, 75, 78

Answer

\(r=0.964\). Hence there is a very strong positive correlation.
Coefficient of determination \(=92.953\%\).
Regression equation: \(\hat{Y}=51.567+1.733X\).

The predicted practical score increases by about \(1.733\) marks for each additional coding hour per week.

2

Curing Time and Compressive Strength

Find Karl Pearson’s correlation coefficient, coefficient of determination and interpretation.

Curing time \(X\) days: 3, 7, 14, 21, 28, 35, 42
Compressive Strength \(Y\) MPa: 18, 24, 29, 32, 35, 37, 39

Answer

\(r=0.964\). Hence there is a very strong positive correlation.
Coefficient of determination \(=92.936\%\).
Regression equation: \(\hat{Y}=19.850+0.500X\).

About \(92.936\%\) of the variation in compressive strength is explained by curing time according to the fitted model.

3

Pavement Deterioration Index and ADT

Find the regression equation of PDI on ADT and estimate the PDI when ADT is 32,000.

PDI \(Y\): 72, 68, 59, 63, 70, 55, 50, 62, 48, 58
ADT \(X\) in '000 vehicles: 12, 18, 22, 16, 14, 25, 28, 20, 30, 24

Answer

Regression equation: \(\hat{Y}=88.082-1.320X\).
For \(X=32\), \(\hat{Y}=45.851\).

The slope is negative. As traffic increases, predicted PDI decreases. Since lower PDI means more deterioration, heavier traffic is associated with more pavement deterioration.

4

CPU Utilization and Power Consumption

Compute Pearson’s correlation coefficient, draw the suitable graph and interpret.

CPU Utilization \(X\): 10, 15, 25, 30, 40, 50, 60, 70, 80, 90
Power Consumption \(Y\): 85, 92, 110, 118, 135, 152, 168, 185, 202, 220

Answer

\(r=0.9999\), approximately \(1\).
Coefficient of determination \(=99.987\%\).
Regression equation: \(\hat{Y}=67.492+1.685X\).

A scatter diagram is suitable. The points show a very strong upward linear pattern. Power consumption increases as CPU utilization increases.

5

Advertising and Sales with Standard Error

Develop the simple regression line to predict sales from advertising expenses. Also find the standard error of estimate and comment.

Advertising \(X\): 12.5, 3.7, 21.6, 60.0, 37.6, 6.1, 16.8, 41.2
Sales \(Y\): 148, 55, 338, 994, 541, 89, 126, 379

Answer

Regression equation: \(\hat{Y}=-46.292+15.240X\).
Standard error of estimate \(=108.758\) million.

The standard error of estimate gives the typical deviation of observed sales from the fitted regression line. Here the fitted values may differ from observed sales by about \(108.758\) million on average.

6

Air Velocity and Evaporation Coefficient

Fit a straight line to the data and construct a 95% confidence interval for the intercept.

Air velocity \(X\): 20, 60, 100, 140, 180, 220, 260, 300, 340, 380
Evaporation coefficient \(Y\): 0.18, 0.37, 0.35, 0.78, 0.56, 0.75, 1.18, 1.36, 1.17, 1.65

Answer

Regression line: \(\hat{Y}=0.069+0.003829X\).
Standard error of estimate \(=0.159\).
95% confidence interval for intercept: \((-0.164,\;0.302)\).

Evaporation coefficient tends to increase as air velocity increases.

7

Beam Load and Maximum Deflection

Calculate the linear regression by treating deflection as dependent variable and estimate maximum deflection when load is 170 kg.

Load \(X\) kg: 100, 120, 140, 160, 180, 200
Deflection \(Y\) cm: 0.45, 0.55, 0.60, 0.70, 0.80, 0.85

Answer

Regression equation: \(\hat{Y}=0.048+0.004071X\).
For \(X=170\), \(\hat{Y}=0.740\) cm.

For each additional kg of load, predicted maximum deflection increases by about \(0.004071\) cm.

8

Production and Export

Plot a suitable graph, find the correlation coefficient between production and export, and interpret.

Production \(X\): 55, 56, 58, 59, 60, 60, 62
Export \(Y\): 35, 38, 38, 39, 44, 43, 45

Answer

\(r=0.932\). Hence there is a strong positive correlation.
Coefficient of determination \(=86.805\%\).
Regression equation: \(\hat{Y}=-43.120+1.424X\).

A scatter diagram is suitable. Export tends to increase as production increases.

9

Advertising and Sales Prediction

Develop the simple regression line to predict sales from advertising expenditure. Predict sales when advertising expenditure is 25 million.

Advertising \(X\): 12.5, 3.7, 21.6, 60, 37.6, 6.1, 16.8, 41.2
Sales \(Y\): 148, 55, 338, 994, 541, 89, 126, 379

Answer

Regression equation: \(\hat{Y}=-46.292+15.240X\).
For \(X=25\), \(\hat{Y}=334.702\) million.

For each additional million spent on advertising, predicted sales increase by about \(15.240\) million according to the fitted model.

10

Large Paired Sample: Mean, Consistency and COD

Find sample mean for each variable, decide which series is more consistent, find standard error of difference of means, and calculate coefficient of determination.

\(X\): 46, 55, 49, 57, 56, 46, 66, 47, 51, 61, 56, 68, 68, 48, 58, 68, 45, 50, 59, 45, 66, 62, 57, 57, 57, 59, 47
\(Y\): 27, 37, 24, 43, 43, 50, 49, 48, 29, 46, 43, 40, 18, 39, 32, 48, 18, 45, 32, 26, 27, 29, 47, 37, 37, 27, 30

Answer

\(ar{X}=55.704,\quad ar{Y}=35.963\)
\(CV_X=13.616\%,\quad CV_Y=26.887\%\)
Standard error of difference of means \(=2.365\)
\(r=0.126,\quad r^2 imes100=1.594\%\)

Since \(CV_X

11

Building Height and Natural Frequency

Compute the regression line of natural frequency on building height and estimate the natural frequency when height is 26.

Height \(X\): 18, 10, 9, 10, 21, 12, 21, 9
Natural frequency \(Y\): 38, 34, 29, 48, 24, 29, 28, 49

Answer

Regression equation: \(\hat{Y}=47.987-0.954X\).
For \(X=26\), \(\hat{Y}=23.193\).

The slope is negative. In this sample, predicted natural frequency decreases as height increases.

12

Calculator-Based Paired Data

Using a scientific calculator, find CV of each variable, decide consistency, find Karl Pearson correlation coefficient, and calculate coefficient of determination.

\(X\): 46, 55, 49, 57, 56, 46, 66, 47, 51, 61, 56, 68, 68, 48, 58, 68, 45, 50, 59, 45, 66, 62, 57, 57, 57, 69, 47
\(Y\): 27, 37, 24, 43, 43, 50, 49, 48, 29, 46, 43, 40, 18, 39, 32, 48, 18, 45, 32, 26, 21, 29, 47, 37, 37, 27, 30

Answer

\(ar{X}=56.074,\quad CV_X=14.240\%\)
\(ar{Y}=35.741,\quad CV_Y=27.835\%\)
\(r=0.044,\quad r^2 imes100=0.196\%\)

Since \(CV_X

13

Starting Salary and Grade Point Average

Develop the estimating equation and estimate the salary when grade point average is 3.75.

Salary \(Y\) in 000: 36, 30, 30, 24, 27, 33, 21, 27
GPA \(X\): 4.0, 3.0, 3.5, 2.0, 3.0, 3.5, 2.5, 2.5

Answer

Regression equation: \(\hat{Y}=9+6.5X\).
For \(X=3.75\), \(\hat{Y}=33.375\).

Since salary is in thousands, the estimated starting salary is \(33.375\) thousand.

14

Uniformity, Relationship and Coefficient of Determination

Using the paired data, find which series is more uniform, find the degree of relationship, and compute coefficient of determination.

\(X\): 75, 96, 80, 72, 93, 71, 86, 72, 79, 98, 77, 84, 70, 96, 85, 84, 94, 71, 89, 92, 76, 88, 78, 83, 96, 87, 71
\(Y\): 66, 97, 76, 64, 77, 89, 73, 75, 85, 92, 81, 85, 90, 87, 59, 93, 67, 64, 94, 77, 66, 62, 92, 56, 58, 82, 75

Answer

\(CV_X=10.992\%,\quad CV_Y=16.151\%\)
So \(X\) is more uniform.
\(r=0.118\). Hence the relationship is very weak positive.
Coefficient of determination \(=1.399\%\).

Only about \(1.399\%\) of the variation in \(Y\) is explained by \(X\) according to the fitted linear model. This is not a strong linear relationship.

Formulae

Available Formula Sheets

⌂ Home
Latest
Latest Content Auto
Home