Casino Size Revenue
1. **Problem:** Determine if there is sufficient evidence of a linear correlation between casino size and revenue, and whether increasing size can increase revenue.
2. **Given Data:**
Size (thousands sq ft): $160, 227, 140, 144, 161, 147, 141$
Revenue (millions): $189, 157, 140, 127, 123, 106, 101$
3. **Step 1: Calculate the linear correlation coefficient $r$**
Formula: $$r = \frac{n\sum xy - \sum x \sum y}{\sqrt{(n\sum x^2 - (\sum x)^2)(n\sum y^2 - (\sum y)^2)}}$$
Calculate sums:
$\sum x = 1120$, $\sum y = 943$, $\sum xy = 153,964$, $\sum x^2 = 181,066$, $\sum y^2 = 139,927$, $n=7$
Calculate numerator:
$$7 \times 153,964 - 1120 \times 943 = 1,077,748 - 1,056,160 = 21,588$$
Calculate denominator:
$$\sqrt{(7 \times 181,066 - 1120^2)(7 \times 139,927 - 943^2)} = \sqrt{(1,267,462 - 1,254,400)(979,489 - 889,249)} = \sqrt{13,062 \times 90,240} \approx \sqrt{1,178,000,000} \approx 34,320$$
Calculate $r$:
$$r = \frac{21,588}{34,320} \approx 0.629$$
Interpretation: $r=0.629$ indicates a moderate positive linear correlation.
4. **Step 2: Linear regression equation $y = mx + b$**
Slope $m$:
$$m = \frac{n\sum xy - \sum x \sum y}{n\sum x^2 - (\sum x)^2} = \frac{21,588}{13,062} \approx 1.653$$
Intercept $b$:
$$b = \frac{\sum y - m \sum x}{n} = \frac{943 - 1.653 \times 1120}{7} = \frac{943 - 1,851.36}{7} = \frac{-908.36}{7} \approx -129.77$$
Regression equation:
$$\hat{y} = 1.653x - 129.77$$
5. **Step 3: Predict revenue for size 200**
$$\hat{y} = 1.653 \times 200 - 129.77 = 330.6 - 129.77 = 200.83$$
6. **Step 4: Accuracy of prediction**
Since $r$ is moderate, prediction has some uncertainty. Also, 200 is within the data range, so prediction is reasonable but not guaranteed.
---
7. **Problem 2a:** Find linear regression equation (already done above):
$$\hat{y} = 1.653x - 129.77$$
8. **Problem 2b:** Predict revenue for size 200 (done above):
$$200.83$$ million dollars approximately.
9. **Problem 3a:** Calculate linear correlation coefficient $r$ for time and height.
Data:
Time $t$: $0.0, 0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, 1.8$
Height $h$: $0.0, 1.7, 3.1, 3.9, 4.5, 4.7, 4.6, 4.1, 3.3, 2.1$
Calculate sums:
$n=10$
$\sum t = 9.0$
$\sum h = 31.0$
$\sum th = 22.68$
$\sum t^2 = 11.4$
$\sum h^2 = 117.88$
Calculate numerator:
$$10 \times 22.68 - 9.0 \times 31.0 = 226.8 - 279 = -52.2$$
Calculate denominator:
$$\sqrt{(10 \times 11.4 - 9.0^2)(10 \times 117.88 - 31.0^2)} = \sqrt{(114 - 81)(1178.8 - 961)} = \sqrt{33 \times 217.8} = \sqrt{7187.4} \approx 84.77$$
Calculate $r$:
$$r = \frac{-52.2}{84.77} \approx -0.616$$
10. **Problem 3b:** Interpretation
$r \approx -0.616$ indicates a moderate negative linear correlation between time and height.
11. **Problem 3c:** Mistake without scatterplot
Without a scatterplot, one might incorrectly assume a linear relationship when the data actually follows a parabolic trajectory (height first increases then decreases), so linear correlation is misleading.
---
**Final answers:**
- Casino size and revenue have moderate positive correlation ($r=0.629$).
- Regression equation: $\hat{y} = 1.653x - 129.77$.
- Predicted revenue for size 200: $200.83$ million.
- Time and height have moderate negative correlation ($r=-0.616$).
- Scatterplot is essential to avoid misinterpretation of nonlinear data.