Stats Flashcards

1
Q

R-squared value

A

The proportion of variation explained by a trendline

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

What type of R-squared is better?

A

higher

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

When faced with competing hypotheses

A

select the one that makes the fewest assumptions

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

Occam’s razor & The Parsimony Principle:

A

when faced with competing hypotheses, select the one that makes fewest assumptions…

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

Residual

A

distance between a fitted line and an observation.

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

SStot

A

total sum of squares

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

SSres

A

sum of squares of residuals

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

Extrapolation

A

extending a relationship into unknown

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

Experiences curves

A

Power law: y= axb
Increasing effort and diminishing rewards suggests power law
Exponent >1 increasing returns
Exponent<1 decreasing returns

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

Pearsons

A

Measure linear trend using Pearson’s product – moment correlation coefficient
Appropriate for normally distributed data
Measures strength of linear trend for paired data point
-1 to 1

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

P value

A

The p value is the likelihood of seeing a pattern at random
If p>0.05 report the exact value – non significant
If 0.05>p>0.01 report the exact value – significant
If p<0.01 don’t report the exact value – significant

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

Five sigma

A

Five-sigma indicates that the observed statistic was more than five standard deviations from the mean.

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

Correlation

A
  • Degrees of freedom: N-2

* Testing a relationship between samples – an association with neither variable fixed by the experimenter

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Regression

A

• Testing a relationship between samples: an association with a variable fixed manipulated by experimenter

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

Two sample t-test

A
  • Testing a difference between samples: two groups where each subject receives one treatment
  • Degrees of freedom: N-2
  • If critical value is more than t value null hypothesis is accepted
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Mann-Whitney U test

A
  • Testing a difference between samples: non-parametric test for difference between two samples
  • When data is not normally distributed
17
Q

Chi squared test

A
  • Testing frequency or count data: frequencies of observations in different categories
  • Degrees of freedom = Number if categories -1
  • f critical value is more than chi squared null hypothesis is rejected