Distribution and probability Flashcards
inference
allows us to make predictions about the population from which the data was selected
what kind of data can have normal distribution
continuous
what kind of data can have binomial distribution
binary data
what kind of data can have poisson distribution
for events occurring at random intervals of time
characteristics of normal distribution
bell shaped single, central peak symmetrical equal mean, median, mode continuous
what two features define normal distribution
mean and variance
how can you assess the normality of distribution
informal review of properties
inspection of a normal plot
Shapiro-Wilk test
why do we transform data
transformed so they are plausibly normally distributed
logarithmic transformation is appropriate for:
- data that is fairly skewed
- groups of data in which the variances are proportional to the mean
square root transformation:
- data that is slightly skewed
- counts
reciprocal transformation:
- data that is highly skewed
cube transformation:
- data relating to volumes
logit transformations
- proportions