Reading 15 Linear Regression with One Regressor| Quantitative Analysis-Part 1| FRM Video Tutorials Video Lecture | Risk Management: An Essential Training - Management

simply learn your pace your place
financial risk manager frm par one or
the frm exam covers of fundamental tools
and techniques used in risk management
and the theories that underlie their use
linear regression with one regressor
after reviewing the basics of
probability and statistics let us now
move on to another important topic
regression analysis this concept has a
wide application in the field of
investment analysis it is basically a
measure of relationship between
variables the next four lectures have
been devoted to different types of
regression and testing their statistical
significance through hypothesis let us
begin from the regression of two
variables agenda the agenda of our
discussion will be learning about the
regression of two variables that is
there will be one independent variable
and one will be a dependent variable
regression will be assumed to be linear
that is the variable changes linearly
with a change in independent variable we
will begin with the concept of
regression analysis then we'll briefly
look at the various assumptions used in
linear regression you will then estimate
the parameter values of the regression
equation and demonstrate it with the
help of an example at the end concept of
regression suppose we have data for two
variables and we need to know how the
dependent variable Y changes with the
independent variable X a scatterplot
depicts the approximate relationship in
our example we can see that there is an
approximate increase in Y as x increases
also the change seems to be linear in
regression analysis our main aim would
be to determine a line that is the best
fit for the scatterplot this means that
the line should be the line that best
describes the linear relationship in
other words we approximate the
parameters of the line the slope and
coefficient in such a way as to minimize
the error terms or the deviation of the
actual values from the predicted values
as you can see not all points lie on the
line so there is an error in this kind
of linear approximation
and it is this error that we need to
minimize the error exists because the
relationship is not exactly linear and
there might be other variables that are
affecting the value of the dependent
variable for example petrol prices do
change with increase in crude oil prices