UNIT 2 AP STATS

demonstrate finding the residual plot with calculator x V: 2,3,7,9,10,10,12,15,16,20 y V: 10,11,13,14,18,20,20,22,22,26

-"stat"-"edit"-put in data-"stat"-"calc"-"linreg(a+bx)"-"enter"-"2nd"-"stat plot"-"plot 1" make sure this is on and where it says y1 should have resid-"2nd"-"mode"-"zoom"-"zoom stat"

explain r

-correlation coefficent -a number between -1 and +1. the closer to 0 the weaker the correlation is. -so r determines the strength of the correlation

influential points (points that effect the linear model)

-points with high or low residuals (outliers) -high leverage point

how to find r^2 on the calculator and define r^2 x V: 2,3,7,9,10,10,12,15,16,20 y V: 10,11,13,14,18,20,20,22,22,26

."stat"-"edit"-put in data-"stat"-"calc"-"linger (a+bx)"-"enter"-"enter" r^2= .93 this is the correlation of determination and it determines the percentage of how reliable the predicter is.

demonstrate finding the best fit line with calculator x V: 2,3,7,9,10,10,12,15,16,20 y V: 10,11,13,14,18,20,20,22,22,26

."stat"-"enter"-put in data-"2nd"-"stat plot"-"enter"-make sure "on" is highlighted-make sure "type" is scatter plot is highlighted-"zoom"-"zoom stat"

when ever a question is,"what effect with removing this outlier have on the regression? Describe how the slope, R^2, and S will change", how do you answer that

.slope will drop .R^2 would increase .S would get smaller

when ever a question is,"what does the value of R^2 say about the model", how do you answer that

According to our linear model, (R ^2) % of the variability in the (y variable) is accounted for by the linear relationship between (y variable) and (x variable).

when ever a question is,"what do we notice in the scatterplot", how do you answer that

C:context D:direction O:outliers F:form S:strength

what is my CDOFS template

The association between (context), is showing a (direction) direction, in a (linear or non linear) format, and showing a (strength) association. And there appears to be ( no or a outlier) outliers.

What is the template for when a problem asks "what does the slope mean in the context"

The predicted (dependent variable), goes ("up" if slope is positive,"down" if slope is negative)(slope number) for every (x variable).

make the regression equation from this data: x V: 2,3,7,9,10,10,12,15,16,20 y V: 10,11,13,14,18,20,20,22,22,26

predicted variable= 8.101+.913(x)

extrapolation

using the info to predict a value that lies well outside the window of the graph. This is DANGROUS