Regression Analysis.

Download Report

Transcript Regression Analysis.

Correlation and Regression
Correlation.
 ‘Correlation’ is a statistical tool which measure the
strength of linear relationship between two
variables.
 It can not measure strong non-linear relationship.
 It can not measure Cause and Effect.
 “Two variables are said to be in correlated if the
change in one of the variables results in a change
in the other variable”.
 Types of Correlation
 There are two important types of correlation.
 (1) Positive and Negative correlation and
 (2) Linear and Non – Linear correlation.
 Positive and Negative Correlation
 If the values of the two variables deviate in the same
direction i.e. if an increase (or decrease) in the values of
one variable results, on an average, in a corresponding
increase (or decrease) in the values of the other variable
the correlation is said to be positive.
 Some examples of series of positive correlation are:
 Heights and weights;
 Household income and expenditure;
 Price and supply of commodities;
 Amount of rainfall and yield of crops.
 Correlation between two variables is said to be negative or
inverse if variables deviate in opposite direction
 Some examples of series of negative correlation are:
 Volume and pressure of perfect gas;
 Current and resistance [keeping the voltage constant] .
 Price and demand of goods etc.
 The Coefficient of Correlation
 It is denoted by ‘r’ which measures the degree of
association between the values of related variables given
in the data set.
 -1≤ r ≤1
 If r >0 variables are said to be positively correlated.
 If r<0 variables are said to be negatively correlated.
 For any data set if r = +1, they are said to be perfectly
correlated positively
 if r = -1 they are said to be perfectly correlated
negatively, and
 if r = 0 they are uncorrelated.
Correlation In Statistica.
 Got to “Statistic” tab and “Basic statistic”.
 Click “Correlation matrix” tab click ok.
 If you want to get correlation only on two variables, go to the “Two
variable list” and select those two variables which are required.
 Click ok and then click on summary.
 If you want to get the correlation matrix for all variables, go
to “Two variable list” tab and click on “select all” tab, click
ok.
 Click the summary , you will get correlation matrix.
 To get scatter plot – click on “Scatter plot of variables” and
then select the variables and click ok.
 You will get the scatter plot as it has came in the right side.
 To get another kind of Scatter plot click on “Graphs” option.
 You will get a series of scatter plot for each and every pairs of
variable.
 The second plot is showing one of those.
Regression
 Regression analysis, in general sense, means the
estimation or prediction of the unknown value of one
variable from the known value of the other variable.
 If two variables are significantly correlated, and if there is
some theoretical basis for doing so, it is possible to predict
values of one variable from the other. This observation leads
to a very important concept known as ‘Regression Analysis’.
 It is specially used in business and economics to study the
relationship between two or more variables that are related
causally and for the estimation of demand and supply
graphs, cost functions, production and consumption
functions and so on.
 Thus, the general purpose of multiple regression is to learn
more about the relationship between several independent or
predictor variables and a dependent or output variable.
 Suppose that the Yield in a chemical process depends on
Temperature and the Catalyst concentration, a multiple
regression that describe this relationship is,
Y=b0+b1*X1+b2*X2+€ → (a)
Where Y = Yield.
X1 = Temp:, X2 = Catalyst cont:.
This is multiple linear regression model with 2 regressors.
 The term linear is used because equation (a) is a linear
function of the unknown parameters bi’s.
Regression Models.

Depending on nature of relationship
regression models are two types.
 Linear regression model, including
a. Simple-linear regression (one indep: var.)
b. Multiple-linear regression.
 Non-Linear regression model, including
a. Polynomial regression.
b. Exponential regression ,etc.
Assumption of Linear regression model.
 The relationship between Y (dependent
variable) and independent variables are
linear.
 The independent variables are mutually
independent to each other.
 The errors are uncorrelated to each
other.
 The error term has fixed variance.
 The errors are Normally distributed.
Data set and Objective.

The current data set has been taken from a chemical
process where we have two input or independent
parameters ,
Temperature and Catalyst feed rate
Response or output parameter : Viscosity of the yield.

1.
2.
3.
Objective.
Establish the linear relation of dependent variable with
independent variables.
Estimate regression coefficients to find out which
variable has significant effect on the response variable.
Check the model adequacy with the help of
assumptions.
Data set.
Assumption of Linearity.
 First of all, as is evident in the name multiple linear
regression, it is assumed that the relationship
between variables is linear.
 In practice this assumption can virtually never be
confirmed; fortunately, multiple regression
procedures are not greatly affected by minor
deviations from this assumption.
 However, as a rule it is prudent to always look at
bivariate scatter plot of the variables of interest.
 If curvature in the relationships is evident, one may
consider either transforming the variables, or
explicitly allowing for nonlinear components.
Scatter plot.
 Go to “Graph” option and select “Scatter plot”.
 Click on “variable” tab and select the variables in the above way.
 Click ok and select the option “Multiple” and “Confidence” .
This will help you to plot multiple graph in a single window.
(If you have large number of variables then plot it separately)
 The scatter plot has
established linear
relationship of the
dependent with
independent
variables.
 Here Viscosity and
Temp are linearly
related to each other,
but Viscosity and
Catalyst concentration
are not.
Parameter Estimation.
 The regression coefficient  (beta) is the average
amount of change in the dependent (either in
positive or negative direction, depending on the sign
of ’s) when the independent changes one unit and
other independents are held constant.
 The b coefficient is the slope of the regression line.
 Intercept (constant,α)- It is the value of dependent
variable when all indep: are set to zero.
 For any independent variable if the corresponding
>0, then that variable is positively correlated with
dependent variable, negatively otherwise.
 OLS (ordinary least squares) is used to estimate
the coefficients in such a way that the sum of the
squared deviations of the distances of all the points
to the line is minimized.
 The confidence interval of the regression coefficient. We
can 95% confident that the real regression coefficient for
the population lies within this intervals.
 If the confidence interval includes 0, then there is no
significant linear relationship between x and y.
 The confidence interval of y – It indicates 95 times out of a
hundred, the true mean of y will be within the confidence
limits around the observed mean of n sampled.
 SEE (Standard error of estimate) is the standard deviation
of the residuals.
In a good model, SEE will be markedly less than the
standard deviation of the dependent variable.
 It can be used to compare the accuracy of different
models, lesser the value better the model.






F-test and P-value: Testing the Overall Significance of the Multiple
Regression Model.
It assume the null hypothesis, H0: b1 = b2 = ... = bk = 0
H1: At least one bi does not equal 0.
If H0 is rejected (if p<.05) we can conclude that,
At least one bi differs from zero.
The regression equation does a better job of predicting the actual
values of y.
t-test: Testing the Significance of a Single Regression Coefficient.
 Is the independent variable xi useful in predicting the actual values
of y ?
For the Individual t-test
H0: bi = 0
H1: bi ≠0
 If H0 is rejected (if p<.05)
The related X has a significant contribution on the dependent
variable,
 R^2 (coefficient of determination)- Is the percent of the
variance in the dependent explained uniquely or jointly by
the independents.
 R-squared can also be interpreted as the proportionate
reduction in error in estimating the dependent when knowing
the independents.
 Adjusted R-Square It is an adjustment of R-square when
one has a large number of independents
 It is possible that R-square will become artificially high
simply because some independent variable "explain" small
parts of the variance of the dependent.
 If there is a huge difference between R-square and Adjusted
R-square then we can assume that some unimportant
independent variables are present in the data set.
 If inclusion of a variable reduces Adjusted R-square it will be
identified as a nonsense parameter for the model.
Estimating coefficients.
 Go to “Statistics” tab and select “Multiple Regression”.
 Select the variables and click ok again click ok.
 Now click “Summary regression results”.
 Left side table showing the model accuracy.
 R-square- Describing the amount of variability that has been
explained by indep: variables, here it is approx. 93%.
 Adjusted R-square – Give an indication whether there is any
insignificant factor or not.
 Adj: R square should be close to Multiple R square, if it is very
smaller than R square then we should go for stepwise regression.
(Adjusted R square always < or = Multiple R square.)
Interpretation of Result.
 Here R square and Adjusted R square are very close to
each other, which indicate a good model.
 In regression analysis R square value will always
increase with the inclusion of parameters , but Adjusted
R square may not be, this indicate the presence of
nuisance parameters in the model.
 The p value for F test is significant (left table) indicate,
there is at least one variable which has significant
contribution to the model.
 The p values for t-test are all significant (as p<.05)(2nd
table) which indicate all these variables has significant
effect on the response.
Multicollinearity.
 Definition Multicollinearity refers to excessive
correlation of the independent variables.
 Ideally independent variable should be uncorrelated to
each other (according to the assumption).
 If the correlation is excessive (some use the rule of
thumb of r >0.90), standard errors of the beta coefficients
become large, making it difficult or impossible to assess
the relative importance of the predictor variables.
 But multicollinearity does not violate OLS assumption, it
still gives unbiased estimate of the coefficient.
Detecting Multicollinearity
 Tolerance The regression of any independent variable on
all the other independents, ignoring the dependent.
As a rule of thumb, if Tolerance ≤ 0.10, a problem with
multicollinearity is indicated.
 VIF (Variance-inflation factor) Is simply the reciprocal of
tolerance.
As a rule thumb, if VIF > 10 , a problem with multicollinearity
is indicated.

C.I (condition indices) Another index for checking
multicollinearity.
As rule thumb , if C.I >30 serious multicollinearity is present
in the data set.
 Some other indication of multicollinearity.
 If none of the t-test for the individual coefficients is
statistically significant, yet the overall F statistic is.
It imply the fact that some coefficients are insignificant
because of multicollinearity.
 Check to see how stable coefficients are when different
samples are used.
For example, you might randomly divide your sample in
two parts. If coefficients differ dramatically, multicollinearity
may be a problem.
 Correlation matrix can also be used to find out which
independent variables are highly correlated (affected by
multicollinearity)
How to perform in Statistica?
 In the “Advanced” tab click on either “Partial correlation” or
“Redundancy” tab.
 You will get the result which contain Tolerance, Partial
correlation, Semi partial correlation etc.
 From the table it is clear that Tolerance > 0.10 , so
Multicollinearity is not a threat for this data set.
Example of VIF
 The data set
contains information
about the physical
and chemical
properties of some
molecules.
 Dependent variablelogP.
 24 numbers of
indep: variables.
 We will first find out
VIF values and also
check the
correlation matrix.
Steps.
 Go to Statistics tab select
“Advanced linear non-linear
model” and click “General
linear model” , select
“Multiple regression”.
 Select variables and click
ok, again click ok.
 Click on “Matrix” tab ,then
select “Partial correlation”.
 The circled variables are
highly affected by
multicollinearity (as
VIF>10).
 Now we can create
correlation matrix to see
which variables are
correlated to each other.
Correlation matrix.
 Go to Statistics tab,
select “Basic
statistics/Tables” then
select “Correlation
matrices”.
 Click on “Two lists”
and select variables.
 Click ok and then
click “ Summary
correlation”.
 The Correlation
Matrix will be
obtained.
 It is clear that large
number of
variables are
highly correlated to
each other and
they are colored as
red, like BO1-X
and DN3 etc.
Methods for Dealing with Multicollinearity.

a.
b.
c.
Several techniques have been proposed for dealing with the
problems of multicollinearity, these are
Collecting additional data: The additional data should be collected
in a manner designed to break up the multicollinearity in the
existing data set. But this method is not always suitable for
economic constraints or for sampling problem.
Variable elimination : If any two or three variables are highly
linearly dependent, eliminating one regressor may be helpful to
reduce multicollinearity . This also may not provide satisfactory
result, since the eliminating variable may have significant
contribution to the predicting power.
Stepwise regression: Most effective method for eliminating
multicollinearity. This method will exclude those variables which
has affected by co linearity step by step and try to maximize the
model accuracy.
Stepwise Regression.
 Stepwise multiple regression, also called statistical
regression, is a way of computing OLS regression in stages.
 First StepThe independent best correlated with the
dependent is included in the equation.
 Second StepThe remaining independent with the highest
partial correlation with the dependent, controlling for the first
independent, is entered.
 This process is repeated, at each stage partialling for
previously-entered independents, until the addition of a
remaining independent does not increase R-squared by a
significant amount (or until all variables are entered, of
course).
 Alternatively, the process can work backward, starting with
all variables and eliminating independents one at a time until
the elimination of one makes a significant difference in Rsquared.
Example of stepwise Regression.
 Go to “Statistics” tab and
select “Multiple
Regression”.
 Select variables ,click
ok.
 Then click on
“Advanced” tab and
select the circled
options.
 Click ok.
 Select the circled option.
(2nd figure)
 Click on “stepwise” tab
and select the circled
option.
 Next click ok.
 You will get the 2nd
window.
 Now you have to click the
tab “Next” until all
important variables are
included into the model.
 Next click on “summary
regression result” to get
the model summary.
Residual Analysis.
 Residuals are the difference between the observed values
and those predicted by the regression equation.
Residuals thus represent error, in most statistical
procedures.
 Residual Analysis is the most important part in Multiple
regression for diagnostic checking of model assumptions.
 Residual analysis is used for three main purposes:
(1) to spot heteroscedasticity (ex., increasing error as the
observed Y value increases),
(2) to spot outliers (influential cases), and
(3) to identify other patterns of error (ex., error associated
with certain ranges of X variables).
Assumptions of Errors.
 The following assumptions on the random errors
are equivalent to the assumptions on the
response variables, which are tested via Residual
Analysis.
(i) The random errors are independent.
(ii) The random errors are normally distributed.
(iii) The random errors have constant variance .
Assumption 3: The Errors are Uncorrelated to each
other. (detection of Autocorrelation).
 Some application of regression involve regressor and
response variables that have a natural sequential order
over time. Such data are called Time series data.
 A characteristic of such data can be that neighboring
observations tend to be somewhat alike. This tendency is
called Autocorrelation.
 Autocorrelation can also be arise in laboratory experiments
,because of the sequence in which experimental runs are
done or drift in instruments calibration .
 Randomization reduce the possibility of Auto correlated
result.
 Parameter estimate may or may not be seriously affected
by Autocorrelation , but autocorrelation will bias the
estimation of variance, and any statistics estimated from
variance like confidence intervals will be wrong.
 How to detect Autocorrelation?
 Various statistical tests are available for detecting
Auto correlation, among them Durbin-Watson test
is widely used method.
 It is denoted by “D”.
If the D value lies between (1.75 , 2.25) residuals
are uncorrelated.
If D <1.75 residuals are correlated, positively and
If D>2.25 residuals are correlated, negatively.
 Go to “Statistics” tab and
select “Multiple regression”,
select variables and click ok
again click ok.
 The 1st window will come ,
click ok ,you will get the 2nd
window.
 Click on “Advanced” tab and
then click on “DurbinWatson statistic”.
 The first column of the
spreadsheet shows the D
statistic value which > 2.25.
 According to rule residuals
are negatively correlated.
 Second column shows the
correlation value.
 Here residuals are
correlated but the
magnitude is not so
high,(<.50) so we can take
this into consideration.
 Analysis done from 1st
dataset. (chemical process
data).
Constant variance of residuals.
 In linear regression analysis, one assumption of the fitted model is that
the standard deviations of the error terms are constant and do not
depend on the x-value.
 Consequently, each probability distribution for y (response variable) has
the same standard deviation regardless of the x-value (predictor).
 This assumption is homoskedasticity.
 How to check this assumption?
 One simple method is to plot the residual values against the fitted
(predicted).
 If this plot shows a systematic pattern then it is fine.
 If it shows abnormal or curvature pattern then there should be problem.
How to manage abnormal condition?
 If the graph shows abnormality , some techniques are there to
manage such condition.
 The usual approach to deal with inequality of variance is to apply
suitable transformation to either independent variables or response
variable.
(Generally transformation of the response are employed stabilize
variance).
 We can also use method of weighted least square instead of
ordinary least square.
 A curved plot may indicate nonlinearity, this could mean that other
regressor variables are needed in the model, for example a squared
term may be necessary.
 A plot of residual against predicted may also reveal one or more
unusually large residuals these points are of course potential outlier.
we can exclude those points from analysis.
 Go to statistics tab and
select “Multiple regression”.
 Select variables and click
ok again click ok.
 You will get the 1st window,
click on
“Residuals/assumptions/pr
ediction” tab and then click
on “perform residual
analysis”.
 You will get the 2nd window,
click on “scatter plot” tab
and click on “predicted v/s
residuals”.
 The above graph showing the predicted v/s residuals scatter plot.
 It is clear that few of points after midpoints are going upward an
downward, which means at that points there are some tendency of
higher residuals positively or negatively.
 In other word that affected points are not predicted properly.
Normality of Residuals.
 In regression analysis last assumption is normality of





residuals.
Small departure from normality assumption do not affect
the model greatly, but gross non normality is potentially
more serious as the t or F test, and confidence intervals
depends on normality assumption.
A very simple method of checking the normality
assumption is to construct “Normal probability plot” of the
residuals.
Small sample size (n<=16) often produce normal plot
that substantially deviate from linearity.
For large sample size (n>=32) the plots are much well
behaved.
Usually at least 20 points or observation are required to
get stable normal probability plot.
Normal probability plot.
 From Multiple
regression go to
residuals tab and
select “perform
residual analysis”.
 Then click on
“probability plot” and
select “Normal plot
of residuals”.
 The normal plot shows
almost good fitting of
normality.
 Small amount of
deviations are there
from the linearity , that
could be overcome
probably if we add
some new
experimental points to
the data set.
(current data contains
only 16 observations).
 Analysis done from 1st
dataset (chemical
process data).
Detecting outlier .
 If any data include some extreme values (outlier)
then it may causes serious problem while checking
the assumptions of regression analysis.
 Few classical techniques are there for detecting
outliers, like Box-Whiskers plot.
 Here we use some residual techniques for
detecting outliers after creating the regression
model.
 One popular method is to check “Standard residual
values” .
 If any value goes beyond ( -3.5 to 3.5),that
particular point will be considered as an outlier.
 From Multiple
regression ,select
variables click ok again
click ok.
 1st window will appear,
select
“Residual/assumption/pr
ediction” then click
“Perform residual
analysis”.
 2nd window will appear
select “Outliers” and
then activate “standard
residuals” and click on
“case wise plot of
outliers”.
 The above output will come.
 Here just check the 5th column (highlighted) ,in this case
all points are within (-3 to +3).
 So no possible outliers are there.
 In the extreme left the case points have demonstrated.
Thank you
Krishnendu Kundu
StatsoftIndia.
Email- [email protected]
Mobile - +919873119520.