Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
38 Cards in this Set
- Front
- Back
What is an Eigenvalue? |
The standard deviation squared + added...? |
|
What is PDF? |
Probability Distribution Function. When integrated over a set of numbers A, will give prob. of X in A. |
|
What is a PMF? |
Probability Mass Function. Probability that a discrete random variable is equal to some real value. |
|
What is a CDF? |
Cumulative Distribution Function. Probability that a random variable is less than or equal to a certain real number. |
|
Describe Central Limits Theorem. |
Given a large sample size, the distribution of means will be normal. Std error works b/c of Theorem. As we approach infinity, looks normal. |
|
What is a random variable? |
Variable independent from samples what we assign any number or you don't know the number. |
|
What is meant by 'power'? Name the 5 components. |
1-beta. 5 Components: alpha, beta, delta/effect size, n/sample |
|
2 Rules of AIC |
1. Choose lowest value. 2. If within 2 units, use less parameters |
|
Covariance vs correlation |
Correlation is standardized covariance. Covariance is a measure of correlation. |
|
What is a linear model? |
Predicting variable by multiplying parameters. |
|
Anova vs Ancova |
Anova: treating parameters of interest as last variable Ancova: parameters are NOT multiplied by each other |
|
What are the parameters of binomial, poisson, uniform, and normal distributions? |
Binomial: n (sample), p, q Poisson: lambda (mean) Uniform: a (min), b (max) Normal: Mew (mean), Sigma (s.d) |
|
Definition of parameter. |
Describes population. Truth. You never know parameters true values of full population. |
|
Definition of statistic. |
Estimate is parameters. Ex: slope, correlation, t-score |
|
What is an anova test? |
Only tells if groups are different. Need to do post-hoc test. Tukey HSD most common. |
|
T-test equation (hypothesis testing)? |
[(delta)mean-theta]/std error |
|
What correction to use for data? |
Bonferroni: more conservative Holm: more liberal |
|
What describes shape of distributions: t, F, chi-squared |
t: df F: df1, df2 Chi-squared: df |
|
What parameter describes chi-squared distribution? Draw it. |
Degrees of freedom. More df = steeper distribution. |
|
Describe F distribution. |
Parameters: df1, df2. Looks like ch-squared distribution. Approaches infinity, looks normal. |
|
Definition of p-value |
Probability of finding observed results if null were true. How valuable is the data. |
|
What are standardized beta values? |
Slope: 2 parameters measured in different values. Standardized slopes to see which slopes are important. |
|
Definitions of mean, median, and mode. |
Mean: average Median: middle Mode: most |
|
Definitions of standard deviation and range. |
Standard deviation: Avg spread of data points around the mean. Also determines how wide your curve is. Range: difference btwn largest and smallest values |
|
Binomial: Parameters, Dis or Con?, Type of data? |
Parameters: n, p, q Discrete Given a probability of something happening, is this gonna happen? |
|
Poisson: parameters, Dis or Con?, Type of data? |
Parameters: lambda (mean) Discrete Concerned w/ numbers or counts. Used for rare events. |
|
Uniform: parameters, Dis or Con?, Type of data? |
Parameters: a, b (min, max) Can be continuous (cdf) or discrete (pmf) Constant probability. |
|
Normal: parameters, dis or con?, Type of data? |
Parameters: mew, sigma/stdev Continuous Data that has a mean and standard deviation. Bell-shaped curve |
|
How do you calculate standard error? |
Stdev or sigma/sqrt(sample size) |
|
How do you calculate the 95% Confidence Internal? |
Mean +/- 1.96*(stdev) |
|
How do you calculate the coefficient of variation? |
(Stdev/mean)*100=% |
|
How do you calculate standard deviation? |
Sqrt((1/N) Epsilon (xi-mew)^2) |
|
How do you calculate variation? |
Average squared deviation from the mean |
|
How are t, F, and chi-squared distributions related? |
Uses df for distribution shape. Both analyze variance. |
|
Difference between binomial and poisson glm. |
Poisson is predicting occurrence based on predictor level. Binomial is predicting events based on predictor level. |
|
When to use factorial anova? |
To compare means of 2 groups |
|
When to use random effects? |
If we think levels we observe in that group too be samples from a larger population |
|
When to use go glm? |
When there's a relationship between variables and accounts for error. |