In: Statistics and Probability
Forecasts the scores you think you will achieve for Weeks, 1, 2, 3, and 4 and send them to your professor.
During Week 5, using your actual scores for Weeks 1, 2, 3, and 4 and your forecast values, calculate a regression equation, r, and R2 for this data.
Week 1-4
Actual score Forecast score
25 25
22 25
35 35
35 35
1. What data did you use in your regression equation? Which is the
independent and dependent variable?
2. What is the regression equation, r, and R2 for your data?
3. Statistically, what does r and R2 tell you?
4. Based on your empirical evidence, how who well did you forecast
your results? Justify your response.
1)
We have used actual score and forecast score in regression equation.
Indepednent = actual score
Dependent = forecast score
2)
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 117 | 120 | 136.75 | 100.0 | 115.00 |
mean | 29.25 | 30.00 | SSxx | SSyy | SSxy |
sample size , n = 4
here, x̅ = Σx / n= 29.25 ,
ȳ = Σy/n = 30.00
SSxx = Σ(x-x̅)² = 136.7500
SSxy= Σ(x-x̅)(y-ȳ) = 115.0
estimated slope , ß1 = SSxy/SSxx = 115.0
/ 136.750 = 0.84
intercept, ß0 = y̅-ß1* x̄ =
5.40
so, regression line is Ŷ =
5.40 + 0.84 *x
SSE= (SSxx * SSyy - SS²xy)/SSxx =
3.291
std error ,Se = √(SSE/(n-2)) =
1.283
correlation coefficient , r = Sxy/√(Sx.Sy)
= 0.98
R² = (Sxy)²/(Sx.Sy) = 0.9671
3)
R tells how strong and direction of the relationship
R sqaure tell how much variation in y can be explained by x
4)
Results are very well forecast results as we can the residuals are almost nthing.
THANKS
revert back for doubt
please upvote