Transcript Sample Slide Heading Image

Primer on Statistics
for Interventional
Cardiologists
Giuseppe Sangiorgi, MD
Pierfrancesco Agostoni, MD
Giuseppe Biondi-Zoccai, MD
What you will learn
•
•
•
•
•
•
•
•
•
•
•
•
Introduction
Basics
Descriptive statistics
Probability distributions
Inferential statistics
Finding differences in mean between two groups
Finding differences in mean between more than 2 groups
Linear regression and correlation for bivariate analysis
Analysis of categorical data (contingency tables)
Analysis of time-to-event data (survival analysis)
Advanced statistics at a glance
Conclusions and take home messages
What you will learn
• Probability distributions
– what is it and what is it for
– discrete: binomial, Poisson
– continuous: normal, Chi-square, F and t
– central limit theorem
What you will learn
• Probability distributions
– what is it and what is it for
– discrete: binomial, Poisson
– continuous: normal, Chi-square, F and t
– central limit theorem
What is a probability distribution?
What is a probability distribution?
Probability distribution: definition
• It identifies either the probability of each value of an
unidentified random variable (for discrete
variables), or the probability of the value falling
within a particular interval (for continuous variables)
• The probability function describes the range of
possible values that a random variable can attain
and the probability that the value of the random
variable is within any (measurable) subset of that
range
• More roughly, a probability distribution is the
universe of all possible cases for a given variable
or function
Probability distribution: definition
• There are thus discrete probability distributions,
when their cumulative distribution function only
increases in jumps. More precisely, a probability
distribution is discrete if there is a finite or
countable set whose probability is 1.
• Otherwhise, probability distributions are called
continuous if their cumulative distribution
function is continuous, which means that it
belongs to a random variable X.
Probability distribution: what for?
• Probability distributions are powerful tools which
are routinely used (either explictly or implicitly)
for making statistical inferences
• It is pivotal to identify the most appropriate
distribution to be exploited for each given
biostatistical problem
• Should you really be concerned?
Probability distribution: what for?
• Probability distributions are powerful tools which
are routinely used (either explictly or implicitly)
for making statistical inferences
• It is pivotal to identify the most appropriate
distribution to be exploited for each given
biostatistical problem
• Should you really be concerned?
…
• Actually no, because when you correctly identify
a given statistical test, you by default choose its
corresponding probability distribution
What you will learn
• Probability distributions
– what is it and what is it for
– discrete: binomial, Poisson
– continuous: normal, Chi-square, F and t
– central limit theorem
Binomial distribution
• The binomial distribution is the discrete
probability distribution of the number of
successes in a sequence of n independent
yes/no experiments, each of which yields
success with probability p
Binomial distribution
• The binomial distribution and the
corresponding binomial test are seldom
used in clinical research, but they are the
most basic example of probability
distribution
• But, how can I recognize a biased die?
• Using the binomial distribution:
I roll the dice 40 times, and compare my
results to the results expected by the binomial
model with n = 40 and p = 1/6
Poisson distribution
• The Poisson distribution is a discrete probability
distribution that expresses the probability of a
number of events occurring in a fixed period of time
if these events occur with a known average rate
and independently of the time since the last event.
The Poisson distribution can also be used for the
number of events in other specified intervals such
as distance, area or volume
Poisson distribution
• The Poisson distribution provides a useful and
efficient way to assess the percentage of time
when a given range of results will be expected.
• You might wish to project a reasonable upper limit
on some event after making a number of
observations.
• Another potential application would be comparing
rates of very rare adverse events, which occur
sparsely in time and space
• The Poisson distribution and the
corresponding tests are however seldom
used in clinical research
What you will learn
• Probability distributions
– What is it and what is it for
– discrete: binomial, Poisson
– continuous: normal, Chi-square, F and t
– central limit theorem
Normal distribution
• The normal distribution, also called the
Gaussian distribution, is an important family of
continuous probability distributions, applicable in
many fields
• Each member of the family may be
defined by two parameters,
location and scale: the mean
("average", μ) and variance
(standard deviation squared, σ2)
respectively
Normal distribution
• The standard normal distribution is the normal
distribution with a mean of zero and a variance
of one
Normal distribution
• The normal distribution is probably the most
powerful tool in biostatistics, with thousand
uses. Why?
– It can be summarized quickly and efficiently by just two
numbers (μ and σ)
– Many probability distributions look normal for large
samples (see central limit theorem)
Chi-square distribution
• Describes the probability distribution of
a random sum (Q) of k independent,
normally distributed random variables
with mean 0 and variance 1
Chi-square distribution
• It is commonly used for chi-square tests for
goodness of fit of an observed distribution to a
theoretical one, and of the independence of two
criteria of classification of qualitative data
• It is a very powerful and robust tool in
biostatistics, second only to the normal
distribution, for comparing categorical
variables and/or goodness of fit
F distribution
• The F distribution is a
continuous probability
distribution
F distribution
• Named by Snedecor as F for Ronald
Aylmer Fisher, is a continuous
probability distribution exploited for
the comparison of continuous
variables
• It is a complex but very potent tool in
biostatistics, and forms the basis of analysis
of variance (ANOVA), as well as many
other complex statistical models and
analyses (eg multivariable linear regression
models)
t distribution
• Student t distribution (or simply the t
distribution) is a probability distribution that
arises in the problem of estimating the
mean of a normally distributed population
when the sample size is small
• Student's distribution arises when (as in
nearly all practical statistical work) the
population standard deviation is unknown
and has to be estimated from the data.
t distribution
Gosset
• Student t distribution (or simply the t
distribution) is a probability distribution that
arises in the problem of estimating the
mean of a normally distributed population
when the sample size is small
• t distribution arises when (as in nearly all
practical statistical work) the population
standard deviation is unknown and has to
be estimated from the data
t distribution
If you look behind
a t distribution,
you will find a…
t distribution
If you look behind
a t distribution,
you will find a…
GUINNESS!!!
t distribution
• The t distribution was developed in 1908 by
William Sealy Gosset, while he worked at a
Guinness Brewery in Dublin, as
he was prohibited from publishing
under his own name. So the
paper was written under the
pseudonym Student
• The t test and the associated
frequentist theory became wellknown through the work of R.A.
Fisher, who called the distribution
“Student's distribution”
t distribution
• The t test is a very useful and friendly test in
biostatistics, probably the most commonly
used one with the chi-square test
t distribution
• The t test is a very useful and friendly test in
biostatistics, probably the most commonly
used one with the chi-square test
What you will learn
• Probability distributions
– what is it and what is it for
– discrete: binomial, Poisson
– continuous: normal, Chi-square, F and t
– central limit theorem
Central limit theorem
• The central limit theorem (CLT) states that the reaveraged sum of a sufficiently large number of
identically distributed independent random
variables each with finite mean and variance will be
approximately normally distributed
• In other words, any sum of many independent
identically distributed random variables will tend to
be distributed according to a particular "attractor
distribution”
• Since many real populations yield distributions with
finite variance (eg weight, height, IQ), this explains
the prevalence of the normal probability distribution
Central limit theorem
Histogram plot of average proportion of
heads in a fair coin toss, over a large
number of sequences of coin tosses.
Central limit theorem
Histogram plot of average proportion of
heads in a fair coin toss, over a large
number of sequences of coin tosses.
In other words, if you collect enough cases, most
variables will be distributed normally around their
means and according their variances, and parametric
statistics and tests will be potentially applicable
Everything is connected –
applications of the CLT
• From binomial to Poisson:
– As n approaches ∞ and p approaches 0 while np
remains fixed at λ > 0 or at least np approaches λ > 0,
then the Binomial (n, p) distribution approaches the
Poisson distribution with expected value λ
• From binomial to normal:
– As n approaches ∞ while p remains fixed, the
distribution of
approaches the normal
distribution with expected value 0 and variance 1
(this is just a specific case of the central limit theorem)
Frequency
When is a distribution normal?
Value
When is a distribution normal?
Testing normality assumptions
Rules of thumb
1. Refer to previous data or analyses
(eg landmark articles, large databases)
2. Inspect tables and graphs (eg outliers, histograms)
3. Check rough equality of mean, median,
mode
4. Perform ad hoc statistical tests
•
•
•
Levene test for equality of variances
Kolmogodorov-Smirnov test
Moses-Shapiro test ...
Short test
Sakurai et al, AJC 2007
Thank you for your attention
For any correspondence:
[email protected]
For further slides on these topics feel
free to visit the metcardio.org website:
http://www.metcardio.org/slides.html