lecture 3: intro to probability and statistics Flashcards
what is statistical causality/causation
one thing will directly cause the other, i.e. cause and effect
note: gold standard to determine causality is specific experimental design + randomised studies
what is correlation
any statistical relationship whether causal or not, indicates a predictive relationship that can be exploited in practice
what is Simpson’s paradox
a phenomenon in which a trend appears in several groups of data but disappears/reverses when these groups are combined
what is a random variable
its possible values are numerical outcomes of a random phenomenon, eg. roll dice
what are the two types of random variables (RV)
discrete and continuous
what is a discrete RV (DRV)
it takes on only a countable number of values
what is the probability distribution of a DRV
probability mass function
what is a continuous RV (CRV)
it takes on an infinite number of possible values in some interval
what is the probability distribution of a CRV
probability density function
what are the 3 most important statistics of a RV
expectation, variance and standard deviation
what is the main difference in formulas between DRV and CRV
for CRV, the formulas are integrals
What is Bayes’ rule
P(y⎮X) P(y) P(X⎮y) = ------------------- P(X) P(y) P(X⎮y) = ------------------ ∑y P(y) P(X⎮y)
parameter estimation look at lecture 3 page 27 to 32
maximum likelihood estimation
what is parametric machine learning
a learning model that summarises data with a set of parameters of fixed size
involves 2 steps:
1. select a from for the function eg. normal distribution
2. learn the coefficients for the function from the training data
what is non-parametric learning
algorithms that do not make strong assumptions about the form of the mapping function, good when you don’t have a lot of data and no prior knowledge
