Suppose a researcher wants to evaluate the effect of cholesterol on the blood pressure. The following data on serum cholesterol (in mg/dL) and systolic blood pressure (in mm/Hg) were obtained for 15 patients to explore the relationship between cholesterol and blood pressure:
| S. No | . Cholesterol (mg/dL) | SBP (mm/Hg) |
| 1 | 300 | 150 |
| 2 | 410 | 270 |
| 3 | 380 | 210 |
| 4 | 530 | 310 |
| 5 | 570 | 350 |
| 6 | 490 | 310 |
| 7 | 340 | 210 |
| 8 | 320 | 150 |
| 9 | 280 | 110 |
| 10 | 550 | 320 |
| 11 | 340 | 220 |
| 12 | 350 | 170 |
| 13 | 410 | 260 |
| 14 | 390 | 230 |
| 15 | 450 | 270 |
(i) Fit a linear regression model using the method of least squares.
(ii) Construct the normal probability plot for the regression model fitted on serum cholesterol and systolic blood pressure.
(iii) Test the significance of the fitted regression model.
Nice — I solved this manually (numeric steps shown) and produced the regression, diagnostics, and the numbers you can use to draw the normal probability plot yourself. --- ## Data summary (n = 15) $$ \begin{aligned} \sum x &=6110,\qquad \sum y =3540,\\ \sum x^2 &=2{,}608{,}100,\qquad \sum xy =1{,}529{,}500. \end{aligned} $$ --- ## (i) Least-squares linear regression We fit the model $y=\beta_0+\beta_1 x + \varepsilon$. Using the standard normal-equation formulas $$ \beta_1=\frac{n\sum xy-(\sum x)(\sum y)}{\,n\sum x^2-(\sum x)^2\,},\qquad \beta_0=\frac{\sum y-\beta_1\sum x}{n}, $$ compute numerator and denominator for $\beta_1$: $$ \begin{aligned} \text{num} &= n\sum xy-(\sum x)(\sum y)=15(1{,}529{,}500)-6110(3540)=1{,}313{,}100,\\ \text{den} &= n\sum x^2-(\sum x)^2=15(2{,}608{,}100)-6110^2=1{,}789{,}400. \end{aligned} $$ So $$ \beta_1=\frac{1{,}313{,}100}{1{,}789{,}400}=0.733821\quad(\text{approx}), $$ $$ \beta_0=\frac{3540-0.733821\cdot6110}{15}=-62.909914\quad(\text{approx}). $$ Estimated regression line $$ \boxed{\ \hat y = -62.9099 \;+\; 0.733821\,x\ } $$ (Interpretation: each 1 mg/dL rise in cholesterol is associated with ≈0.734 mmHg increase in SBP.) --- ## Goodness-of-fit and sums of squares Let $\bar y=236$, $\bar x=407.3333$. Calculated sums of squares: * Total sum of squares: $\displaystyle SST=\sum (y-\bar y)^2 =70{,}560.000$. * Regression sum of squares: $\displaystyle SSR=\sum(\hat y-\bar y)^2 =64{,}238.725$. * Error sum of squares: $\displaystyle SSE=\sum (y-\hat y)^2 =6{,}321.275$. Check: $SST = SSR + SSE$ (70,560 ≈ 64,238.725 + 6,321.275). Coefficient of determination: $$ __________ __________ ______ ___ ______ ________ _________ _______.
________ _______ ____ ________ _______ _____ ________ ____ ________ ___.
____ __________ ___ ___ ________ ___ _____ _______ ____ _______ _______ ___.
_______ ____ _____ ______ ____ ___ _______ _____ ________.
_____ __________ _________ _____ _______ ______ _______ ___ _________ ____ _____ _______.
___ ______ _____ __________ ________.
_____ _____ ___ _______ _______ ________ _______ _________.
__________ _______ _______ ___ ______ ___ _______ ______.
____ ___ _________ ___ _______ ______ _________ ______ _______ ______ ____ _______.
_____ ___ _______ _________ _____ ________ _________ __________ ___ __________ ____.
___ _________ __________ ________ ___ __________ _____ __________ _________ _____.
_______ _____ _________ __________ ________ __________ _______ _______ ______ ______ ________ _________.
_______ _________ _______ ______ ________ ________ _____ ____ ___ ______.
_____ __________ _____ _______ _____ _____ ___ ___ ___ _______.
___ ________ _______ ________ __________ __________ _______ __________ _________ _____ _____.
________ _________ ______ ___ ____ ____ ___.
_____ _____ ________ _______ ________ _____.
_________ ____ _____ ______ ____ _______ ________ ________ __________.
____ ________ _________ ____ ______ _______ __________ ________ ____ ___.
_________ ___ _____ ___ ________ ____.
________ ___ _____ _________ ____ ____.
__________ ___ _____ _______ _____ _________ ___ __________ ______ __________ ________ ___.
____ ________ _______ ___ __________ ________ ____ _______ _____ _______.
__________ ___ _______ _____ ____.
____ _______ ________ _____ ______.
_____ ______ ______ ________ _____ ____ ____ ________ __________ ________.
___ _________ __________ ___ _________ _______ ________ ___ _______ ___ __________ ______.
_________ _____ _________ ______ __________ _______ __________ ______.
___ _______ ____ ________ __________ ________ ______ ___ _______ ___ _________.
____ _____ _________ _____ ____.
__________ _____ _________ _________ _________ _____.
__________ ____ _________ _________ ________ ____ ________ _________.
_________ ____ ____ ______ __________ _________ _____ ___ _____ ______.
__________ ___ _____ _______ ______ ______ ______ ______.
____ __________ __________ ________ ________ ________ _____ ___ ________ ___ ___ ____.
_______ ____ ___ ________ ________ _______ ___ _________.
_____ ________ _________ _______ _______ ____ ______ ______ _________ _____ __________.
________ ______ ___ ________ ______ __________ _____ __________ ____ _______.
_________ _______ __________ ______ __________ _________.
_________ ______ ________ ______ __________ ____ _______ ________ ______ ___.
_______ _________ _____ _______ ___ _______ ________ _______.
_________ _____ ____ _____ ______.
___ ________ __________ ___ __________ ____ ______.
_____ ______ _________ ___ _________ ______ ___ _______ __________ _____.
______ _____ _______ ____ ____ _______ _________ ______ ________.
________ ______ ________ ______ ______.
_____ _________ ______ _________ ______.
_________ _______ ____ ______ __________ _____ ________ ___ _______ ____ ______.
____ _____ _________ _______ ___ ___.
________ _____ __________ ________ __________ __________ _________ ____ ___ _______ _________.
___ __________ ________ ______ ________ ___ _________ _____ ____ ____ _____.
________ ________ ____ ___ ________ ______ _______ ___ _________ _________ ____ __________.
_____ _____ _______ __________ ___ _____ ________ __________.
_____ ______ ____ _________ ____.
______ ________ ______ __________ ___ ____ ___ __________ ___ ___.
_____ ________ ____ __________ _________ ________ ________ ______ __________ ____ ________ ________.
____ _____ ____ ___ ______ _________ _____.
____ _________ ___ ______ ______ __________ __________ __________ ___ ___ _____ _______.
___ _____ ______ ______ _______ _______ _________ ____ ___.
_________ _____ _______ ___ ________.
____ __________ ____ ______ ___ _______ ____ ______ ______ ____ ____.
______ ___ __________ _______ ________ _________ _______ _______ ______ _______.
_________ __________ ______ ____ ___ ____ ______ ________ ____ ______.
____ _________ ________ _______ ___ ___ ________ _________.
_______ __________ _______ __________ _______ _______ __________ _____.
_________ _____ _____ ____ ___ __________ ___ ___ ______ _____.
_____ ____ ________ __________ ________ _______ ________.
Get Full Answer on WhatsApp
For the data given in Question 2(b), obtain the followings:
(i) Diagonal of the hat matrix. Also, check the leverage points, if any.
(ii) Cook’s Distances. Also, verify the influence points, if any.
Suppose a researcher wants to evaluate the effect of cholesterol on the blood pressure. The following data on serum cholesterol (in mg/dL) and systolic blood pressure (in mm/Hg) were obtained for 15 patients to explore the relationship between cholesterol and blood pressure:
| S. No | . Cholesterol (mg/dL) | SBP (mm/Hg) |
| 1 | 300 | 150 |
| 2 | 410 | 270 |
| 3 | 380 | 210 |
| 4 | 530 | 310 |
| 5 | 570 | 350 |
| 6 | 490 | 310 |
| 7 | 340 | 210 |
| 8 | 320 | 150 |
| 9 | 280 | 110 |
| 10 | 550 | 320 |
| 11 | 340 | 220 |
| 12 | 350 | 170 |
| 13 | 410 | 260 |
| 14 | 390 | 230 |
| 15 | 450 | 270 |
A company conducted a study on its employees to see the relationship of several variables with an employee’s IQ. For this purpose, fifteen employees were selected, and an IQ test and five different personality tests were given to them. Each employee’s IQ was recorded along with scores on five tests. The data are shown in the following table
| Employee | Test 1 | Test 1 | Test 3 | Test 4 | Test 5 | IQ |
| 1 | 83 | 80 | 78 | 77 | 67 | 99 |
| 2 | 73 | 85 | 67 | 80 | 63 | 92 |
| 3 | 81 | 80 | 71 | 81 | 68 | 94 |
| 4 | 96 | 86 | 82 | 83 | 56 | 99 |
| 5 | 84 | 73 | 75 | 75 | 68 | 94 |
| 6 | 72 | 74 | 71 | 67 | 59 | 79 |
| 7 | 84 | 79 | 84 | 84 | 69 | 97 |
| 8 | 54 | 86 | 61 | 69 | 53 | 92 |
| 9 | 86 | 85 | 79 | 78 | 76 | 94 |
| 10 | 42 | 71 | 60 | 80 | 56 | 86 |
| 11 | 83 | 72 | 72 | 78 | 74 | 98 |
| 12 | 63 | 86 | 65 | 85 | 56 | 83 |
| 13 | 69 | 76 | 64 | 85 | 61 | 98 |
| 14 | 81 | 84 | 65 | 85 | 64 | 96 |
State whether the following statements are true or false and also give the reason in support of your answer.
(i) We define three indicator variables for an explanatory variable with three categories.
(ii) If the coefficient of determination is 0.833, the number of observations and explanatory variables are 12 and 3, respectively, then the Adjusted R² will be 0.84.
(iii) For a simple regression model fitted on 15 observations, if we have h₁ = 0.37, then it is an indication to trace the leverage point in the regression model.
(iv) In a regression model is not rejected, then the variable X₁ will remain in the model.
(v) The logit link function is log [-log(1-π)].
Explain the assumptions underlying the multiple linear regression model.
Write a short note on the problem of multicollinearity and autocorrelation.
The following data on the diagnosis of coronary heart disease (where 0 indicates absence and 1 indicates presence), serum cholesterol (in mg/dl), and weight (in kg) were obtained for 80 patients to explore the relationship of coronary heart disease with cholesterol and weight:
| S. No. | Serum Cholesterol (mg/dl) | Weight (kg) | Number of Patients having CHD | Total Number of Patients |
| 1 | 420 | 60 | 10 | 20 |
| 2 | 450 | 68 | 15 | 30 |
| 3 | 400 | 54 | 4 | 15 |
| 4 | 510 | 74 | 2 | 10 |
| 5 | 480 | 62 | 1 | 5 |
(i) Fit a multiple logistic model for the dependence of coronary heart disease on the average serum cholesterol and weight considering as the initial values of the parameters (solve only for one Iteration).