Week 8 Flashcards
Concepts of statistics
Define statistics
a science that studies the collection, analysis, interpretation, presentation and organisation of data
The population is
the total set of observations of a group we study (N)
A sample is
a set of observations taken from the total population (n)
Variables are
things we measure, control or manipulate
Measures of central tendency (location)
mean, mode and median
Measure of spread (dispersion)
Range, quartiles, variance and standard deviation
Is the variance an unbiased estimator of the population?
No, the sample range tends to underestimate the population range
Standard deviation is
the square root of the population variance
Define probability density function
a function that describes the relative likelihood for a random variable to taken on a given value
TRUE OR FALSE
- The PDF is non-positive everywhere?
- The integral over the entire space is equal to 0?
- False, it is non-negative everywhere
2. False, it is equal to 1
What is a uniform distribution?
A family of symmetric probability distributions. Can be continuous or discrete data
Which 2 parameters are normal distribution defined by?
Location (mean) and scale (variance)
What are the parameters of a standard normal distribution?
Mean = 0 Variance = 1
Define accuracy, precision and bias
Acc - how close measurements are to the true value trying to be measured.
Prec - how repeatable the measurement is within itself
Bias - a systematic lack of accuracy in which all data deviates from true value in same direction
What is the purpose of a statistical test?
To evaluate the relationship between variables
What does a statistical test give us?
a significance level (p-value) that tells us the probability of error in rejecting the null hypothesis
Define statistical significance
probability that observed relationship or difference in a sample occurred by pure chance
What does the p-value represent?
the probability of error that is involved in rejecting the null hypothesis based on our observed data
Parametric vs non-parametric
Parametric tests make assumptions about the distribution of data i.e. normally distributed.
Non-parametric tests do not
Paired vs non-paired
Paired tests have data that has been observed from a particular individual at different time points. Non-paired data is observed from different individuals.
One-tailed vs two-tailed
One-tailed only want to know about a single direction of outcome in the data where two-tailed looks at all possible outcomes