Given:
\(\displaystyle{n}=\ \text{Sample size}\ ={30}\)
a) Fat is on the horizontal axis and Death rate is on the vertical axis.
b) When there is no strong curvature presents in the scatterplot, then it is safe to assume that there is a linear relationship between the variables and thus it is then reasonable to find a regression line.
We note that the scatterplot of part (a) does not contain strong curvature and thus it is reasonable to find the regression line.
c) We determine all necessary sums:
\(\displaystyle\sum\ {x}_{{{i}}}={2911}\)

\(\displaystyle\sum\ {y}_{{{i}}}={298.1}\)

\(\displaystyle\sum\ {x}_{{{i}}}\ {y}_{{{i}}}={33630.1}\)

\(\displaystyle\sum\ {{x}_{{{i}}}^{{{2}}}}={323963}\)

\(\displaystyle\sum\ {{y}_{{{i}}}^{{{2}}}}={3642.85}\) Next, we can determine \(\displaystyle{S}_{{\times}}\ \text{and}\ {S}_{{{x}{y}}}\)

\(\displaystyle{S}_{{\times}}=\ \sum\ {{x}_{{{i}}}^{{{2}}}}\ -\ {\frac{{{\left(\sum\ {x}_{{{i}}}\right)}^{{{2}}}}}{{{n}}}}={323963}\ -\ {\frac{{{2911}^{{{2}}}}}{{{30}}}}={41498.9667}\)

\(\displaystyle{S}_{{{x}{y}}}=\ \sum\ {x}_{{{i}}}\ {y}_{{{i}}}\ -\ {\frac{{{\left(\sum\ {x}_{{{i}}}\right)}{\left(\sum\ {y}_{{{i}}}\right)}}}{{{n}}}}={33630.1}\ -\ {\frac{{{2911}\ \cdot\ {298.1}}}{{{30}}}}={4704.4633}\) The estimate b of the slope \(\displaystyle\beta\ \text{is the ratio of}\ {S}_{{{x}{y}}}\ \text{and}\ {S}_{{\times}}:\)

\(\displaystyle{b}=\ {\frac{{{S}_{{{x}{y}}}}}{{{S}_{{\times}}}}}=\ {\frac{{{4704.4633}}}{{{41498.9667}}}}={0.1134}\) The mean is the sum of all values divided by the number of values: \(\displaystyle\overline{{{x}}}=\ {\frac{{\sum\ {x}_{{{i}}}}}{{{n}}}}=\ {\frac{{{2911}}}{{{30}}}}={97.0333}\)

\(\displaystyle\overline{{{y}}}=\ {\frac{{\sum\ {y}_{{{i}}}}}{{{n}}}}=\ {\frac{{{298.1}}}{{{30}}}}={9.9367}\) The estimate a of the intercept \(\displaystyle\alpha\) is the average of y decreased by the product of the estimate of the slope and the average of x. \(\displaystyle{a}=\ \overline{{{y}}}\ -\ {b}\ \overline{{{x}}}={9.9367}\ -\ {0.1134}\ \cdot\ {97.0333}=\ -{1.0634}\) Generel least-squares equation: \(\displaystyle\hat{{{y}}}=\ \alpha\ +\ \beta\ {x}.\ \text{Replace}\ \alpha\ \text{by}\ {a}=\ -{1.0634}\ \text{and}\ \beta\ \text{by}\ {b}={0.1134}\ \text{in the general least-squares equation:}\)

\(\displaystyle\hat{{{y}}}={a}\ +\ {b}{x}=\ -{1.0634}\ +\ {0.1134}{x}\) d) Let us evalute the regression line in part (c) at \(\displaystyle{x}={92}:\)

\(\displaystyle\hat{{{y}}}=\ -{1.0634}\ +\ {0.1134}{\left({92}\right)}={9.3661}\) Thus the predicted death rate for a nation with a capita fat consumption of 92 grams per day is 9.3661 deaths per 100,000 males. e) We determine all necessary sums: \(\displaystyle\sum\ {x}_{{{i}}}={2911}\)

\(\displaystyle\sum\ {y}_{{{i}}}={298.1}\)

\(\displaystyle\sum\ {x}_{{{i}}}\ {y}_{{{i}}}={33630.1}\)

\(\displaystyle\sum\ {{x}_{{{i}}}^{{{2}}}}={323963}\)

\(\displaystyle\sum\ {{y}_{{{i}}}^{{{2}}}}={3642.85}\) Determine the correlation coefficient: \(\displaystyle{r}=\ {\frac{{\sum\ {x}_{{{i}}}\ {y}_{{{i}}}\ -\ {\left(\sum\ {x}_{{{i}}}\right)}\frac{{\sum\ {y}_{{{i}}}}}{{n}}}}{{\sqrt{{{\left[\sum\ {{x}_{{{i}}}^{{{2}}}}\ -\ \frac{{\left(\sum\ {x}_{{{i}}}\right)}^{{{2}}}}{{n}}\right]}{\left[\sum\ {{y}_{{{i}}}^{{{2}}}}\ -\ \frac{{\left(\sum\ {y}_{{{i}}}\right)}^{{{2}}}}{{n}}\right]}}}}}}\)

\(\displaystyle=\ {\frac{{{3363.1}-\ {\left({2911}\right)}\frac{{{298.1}}}{{30}}}}{{\sqrt{{{\left[{323963}\ -\ \frac{{2911}^{{{2}}}}{{30}}\right]}{\left[{3642.85}\ -\ \frac{{298.1}^{{{2}}}}{{30}}\right]}}}}}}\)

\(\displaystyle\approx{0.8851}\) If r is positive, then there is a positive linear relationship. If r is negative, then there is a negative linear relationship. If \(\displaystyle{0}\ {<}\ {\left|{r}\right|}\ {<}\ {0.5},\ \text{then the linear relationship is weak. If}\ {0.5}\ {<}\ {\left|{r}\right|}\ {<}\ {0.8},\ \text{then the linear relationship is moderate. If}\ {0.8}\ {<}\ {\left|{r}\right|}\ {<}\ {1},\)</span> then the linear relationship is strong. We note that the linear correlation coefficient r is greater than 0.8 in absolute value and positive, thus there is a strong positive linear relationship between the variables. f) There are no points in the scatterplot that deviate strongly from the general pattern in the scatterplot of part (a) and thus there appear to be no outliers. Since there are no outliers, there are no influential observations either.

\(\displaystyle\sum\ {y}_{{{i}}}={298.1}\)

\(\displaystyle\sum\ {x}_{{{i}}}\ {y}_{{{i}}}={33630.1}\)

\(\displaystyle\sum\ {{x}_{{{i}}}^{{{2}}}}={323963}\)

\(\displaystyle\sum\ {{y}_{{{i}}}^{{{2}}}}={3642.85}\) Next, we can determine \(\displaystyle{S}_{{\times}}\ \text{and}\ {S}_{{{x}{y}}}\)

\(\displaystyle{S}_{{\times}}=\ \sum\ {{x}_{{{i}}}^{{{2}}}}\ -\ {\frac{{{\left(\sum\ {x}_{{{i}}}\right)}^{{{2}}}}}{{{n}}}}={323963}\ -\ {\frac{{{2911}^{{{2}}}}}{{{30}}}}={41498.9667}\)

\(\displaystyle{S}_{{{x}{y}}}=\ \sum\ {x}_{{{i}}}\ {y}_{{{i}}}\ -\ {\frac{{{\left(\sum\ {x}_{{{i}}}\right)}{\left(\sum\ {y}_{{{i}}}\right)}}}{{{n}}}}={33630.1}\ -\ {\frac{{{2911}\ \cdot\ {298.1}}}{{{30}}}}={4704.4633}\) The estimate b of the slope \(\displaystyle\beta\ \text{is the ratio of}\ {S}_{{{x}{y}}}\ \text{and}\ {S}_{{\times}}:\)

\(\displaystyle{b}=\ {\frac{{{S}_{{{x}{y}}}}}{{{S}_{{\times}}}}}=\ {\frac{{{4704.4633}}}{{{41498.9667}}}}={0.1134}\) The mean is the sum of all values divided by the number of values: \(\displaystyle\overline{{{x}}}=\ {\frac{{\sum\ {x}_{{{i}}}}}{{{n}}}}=\ {\frac{{{2911}}}{{{30}}}}={97.0333}\)

\(\displaystyle\overline{{{y}}}=\ {\frac{{\sum\ {y}_{{{i}}}}}{{{n}}}}=\ {\frac{{{298.1}}}{{{30}}}}={9.9367}\) The estimate a of the intercept \(\displaystyle\alpha\) is the average of y decreased by the product of the estimate of the slope and the average of x. \(\displaystyle{a}=\ \overline{{{y}}}\ -\ {b}\ \overline{{{x}}}={9.9367}\ -\ {0.1134}\ \cdot\ {97.0333}=\ -{1.0634}\) Generel least-squares equation: \(\displaystyle\hat{{{y}}}=\ \alpha\ +\ \beta\ {x}.\ \text{Replace}\ \alpha\ \text{by}\ {a}=\ -{1.0634}\ \text{and}\ \beta\ \text{by}\ {b}={0.1134}\ \text{in the general least-squares equation:}\)

\(\displaystyle\hat{{{y}}}={a}\ +\ {b}{x}=\ -{1.0634}\ +\ {0.1134}{x}\) d) Let us evalute the regression line in part (c) at \(\displaystyle{x}={92}:\)

\(\displaystyle\hat{{{y}}}=\ -{1.0634}\ +\ {0.1134}{\left({92}\right)}={9.3661}\) Thus the predicted death rate for a nation with a capita fat consumption of 92 grams per day is 9.3661 deaths per 100,000 males. e) We determine all necessary sums: \(\displaystyle\sum\ {x}_{{{i}}}={2911}\)

\(\displaystyle\sum\ {y}_{{{i}}}={298.1}\)

\(\displaystyle\sum\ {x}_{{{i}}}\ {y}_{{{i}}}={33630.1}\)

\(\displaystyle\sum\ {{x}_{{{i}}}^{{{2}}}}={323963}\)

\(\displaystyle\sum\ {{y}_{{{i}}}^{{{2}}}}={3642.85}\) Determine the correlation coefficient: \(\displaystyle{r}=\ {\frac{{\sum\ {x}_{{{i}}}\ {y}_{{{i}}}\ -\ {\left(\sum\ {x}_{{{i}}}\right)}\frac{{\sum\ {y}_{{{i}}}}}{{n}}}}{{\sqrt{{{\left[\sum\ {{x}_{{{i}}}^{{{2}}}}\ -\ \frac{{\left(\sum\ {x}_{{{i}}}\right)}^{{{2}}}}{{n}}\right]}{\left[\sum\ {{y}_{{{i}}}^{{{2}}}}\ -\ \frac{{\left(\sum\ {y}_{{{i}}}\right)}^{{{2}}}}{{n}}\right]}}}}}}\)

\(\displaystyle=\ {\frac{{{3363.1}-\ {\left({2911}\right)}\frac{{{298.1}}}{{30}}}}{{\sqrt{{{\left[{323963}\ -\ \frac{{2911}^{{{2}}}}{{30}}\right]}{\left[{3642.85}\ -\ \frac{{298.1}^{{{2}}}}{{30}}\right]}}}}}}\)

\(\displaystyle\approx{0.8851}\) If r is positive, then there is a positive linear relationship. If r is negative, then there is a negative linear relationship. If \(\displaystyle{0}\ {<}\ {\left|{r}\right|}\ {<}\ {0.5},\ \text{then the linear relationship is weak. If}\ {0.5}\ {<}\ {\left|{r}\right|}\ {<}\ {0.8},\ \text{then the linear relationship is moderate. If}\ {0.8}\ {<}\ {\left|{r}\right|}\ {<}\ {1},\)</span> then the linear relationship is strong. We note that the linear correlation coefficient r is greater than 0.8 in absolute value and positive, thus there is a strong positive linear relationship between the variables. f) There are no points in the scatterplot that deviate strongly from the general pattern in the scatterplot of part (a) and thus there appear to be no outliers. Since there are no outliers, there are no influential observations either.