Stats midterm

5.0(1)
studied byStudied by 13 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/38

encourage image

There's no tags or description

Looks like no tags are added yet.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

39 Terms

1
New cards

for MEDIAN: if a+bxi, how will the median change?

a+bmedian

2
New cards

for MEAN: if a+bxi, how will the mean change?

a+bmean

3
New cards

for VARIABILITY: if a+bxi, how will the variability change?

b²σ²

4
New cards

for STANDARD DEVIATION: if a+bxi, how will the standard dev change?

5
New cards

for RANGE: if a+bxi, how will the range change?

b(Range)

6
New cards

for CORRELATION COEFFICIENT: if a+bxi, how will r change?

doesn’t changeif a constant is added, r remains the same.

7
New cards

whats the sum of all mean deviations equal to?

0

<p>0</p>
8
New cards

what are the formulas for VARIABILITY?

σ(variability); σ(standard deviance); CV (coefficient of variance)

9
New cards

what is the min value for the variability and what does it represent?

0, all the variables are the same

10
New cards

what is the min value for the standard deviation and what does it represent?

0, because the standard dev is always positive; ) means that all the values are spread equally and that there is no variation between the data points

11
New cards

what does a higher value for the standard deviation represent?

that the variables are dispersed more

12
New cards

for the GINI COEFFICIENT, how do you calculate Ai, Pi, and Qi?

knowt flashcard image
13
New cards

what are the min and max of the GINI COEFFICIENT, and what do they imply?

min: 0, perfect equality; max:1, perfect inequality

14
New cards

what is heterogeneity? when do e1/e2 imply heterogeneity?

all the categories have the same frequency; occurs at min, 0

15
New cards

what is homogeneity? when do e1/e2 imply homogeneity?

all the variables fall into the same category (are the same); at their respective max

16
New cards

how do you calculate the column conditional relative frequency? in what case does it have to be equal to the marginal relative frequency?

it equals it when there is independence between Y and X; all row+column relative freq should equal the marginal relative freq in order for the 2 variables to be independent

<p>it equals it when there is independence between Y and X; all row+column relative freq should equal the marginal relative freq in order for the 2 variables to be independent</p>
17
New cards

how do you calculate the row conditional relative frequency? in what case does it have to be equal to the marginal relative frequency?

it equals it when there is independence between Y and X; all row+column relative freq should equal the marginal relative freq in order for the 2 variables to be independent

<p>it equals it when there is independence between Y and X; all row+column relative freq should equal the marginal relative freq in order for the 2 variables to be independent</p>
18
New cards

what does 𝑛̂ij represent?

the expected frequency if Y and X would be independent

19
New cards

what formulas are used for INDEPENDENCE? what is the relative measure (most generally used)?

X², Φ; Cramer’s V is more general

20
New cards

what does Φ measure? what are its min and max and what do they represent?

the strenght of association between Y and x; 0, independence; max, dependence

21
New cards

what are the min and max for Cramer’s V and what do they represent?

0, independence; 1, dependence

22
New cards

for mean dependence, how do you compute μyi? but μy?

knowt flashcard image
23
New cards

what does the min and max values for the correlation ratio mean? what does a higher value mean?

0, independence; 1, dependence; the higher the value, he higher the dissimilarity from the mean

24
New cards

what is the extended formula for the deviance? (Dy=De+Dr)

knowt flashcard image
25
New cards

if you plug in the means of x and y into a regression model, what do you get?

the observed values

26
New cards

if you plug in the standard values a regression model, what do you get?

the predicted values

27
New cards

how do you compute the prediction errors\residuals?

knowt flashcard image
28
New cards

what is the sum of the prediction errors\residuals?

0

29
New cards

where do the regression models intersect?

(μx, μy)

30
New cards

what are the alternative formulas for b1?

knowt flashcard image
31
New cards

what is the formula for the least squares method (regression line)?

knowt flashcard image
32
New cards

what does r² measure?

measures the proportion of the total variability of y explained by the regression model

33
New cards

what do 0 and 1 mean for r²?

0, all the points lie on a straight line; 1, really dispersed

34
New cards

when does r imply a negative correlation? when is the mean negative (regarding the correlation between X and Y)?

-1≤ r <0 ; the mean is negative when the correlation/slope is negative

35
New cards

when does r imply a positive correlation? when is the mean positive (regarding the correlation between X and Y)?

0 < r ≤ 1 ; the mean is positive when the correlation/slope is positive

36
New cards

what does it mean when r = 0?

r is null

37
New cards

do we take r as its absolute value?

yes!!!!?!??!?

38
New cards

how do you compute the quartiles? general formula

knowt flashcard image
39
New cards

how do you compute the interquartile range? what is its property? what happens to it if a+bxi?

Q3-Q1 ≥ 0 ; b(Q3-Q1)