Unit 2 Flashcards

1
Q

A good regression line makes the residulas ____\

A

as small as possible

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Least-squares regression line meaning

A

the line that makes the sum of the squared rsiduals as small as possible
##Footnote

goal is to add all the residuals and get =0, as the positive and negative reisduals will cnacel each other out. But even bad fitting lines get 0.50 so to avoid the flaw we square the residulas befor adding & the best line is LSR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Formula for calculationg the sloep fo the LSRL…

A

m = r * (Sy / Sx)
##Footnote

where “r” is the correlation coefficient, “Sy” is the standard deviation of the y values, and “Sx” is the standard deviation of the x values.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Regression to the mean meaning…

A

“Level out”, Find “true” values, “regress t o average”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Formula to caluacluate the y-interecept in the LSRL

A

b = ȳ - m * x̄
##Footnote

where “b” is the y-intercept, “ȳ” is the mean of the y-values, “m” is the slope, and “x̄” is the mean of the x-values.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

residual plot meaning

A

a scatterplot that displays the residuals on the vertical axis and the values of the explanatory variables (or predicted values) on the horizontal axis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

what is the purpose of a residual plot?

A

to help show if the regression line is the best fit; if it helps with the form

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Hoq so you interpret a residual plot

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How do you know when a line is “fit” for a relationship in the regression plot?

A

By the lack of a leftover curved pattern, has random scatterinsert pic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

If the regression plot has a leftover curved pattern in the residual plot what should we do?

A

Consider using a regression model with different form; aka not linear

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Coefficient of determination r^2 definition..

A

measures the proportion (or percentage) of variation in the response variable that is accounted for (or explained) by the explanatory vairable in the linear model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

the standard deviation of the residuals s, defintion…

A

the typical distance between actual y-values and predicted y-values

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Interpreting s points

A
  • mention the actual value of the response variable w.context
  • the s or how much away the response variable is
  • mention of the predicted value w.context
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Interpreting s; The actual ____\ is tupically about ____\ away from the predicted value by the LSRL w/ x= ____\

A

The actual (response variable) is typically about (s) away from the predicted valye by the LSRL w/ x= (explanatory variable)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

high leverage point

A

points with high leverage in regression have much larger or much xmaller x-values than other points in data set

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

outlier

A

in regression the point that does not follow the pattern of the data and has a large residual

17
Q

influential point

A

in regression any point that if removed substanially changes the slope, y intercept, correlation, coefficient of determination, or standard deviaiton of the residuals

18
Q

which model of a residual plot that models the relationship between x and y is the best?

A

the one which…
- has the most random scatter
- if more than 1 has arandom scatter choose mdoel w/ largest coefficent of r^2