5. Regression Model

Loading...

En provenance du cours de University of London

Statistics for International Business

96 notes

University of London

Statistics for International Business

96 notes

Cours 4 sur 6 dans Specialization International Business Essentials

This course introduces core areas of statistics that will be useful in business and for several MBA modules. It covers a variety of ways to present data, probability, and statistical estimation. You can test your understanding as you progress, while more advanced content is available if you want to push yourself. This course forms part of a specialisation from the University of London designed to help you develop and build the essential business, academic, and cultural skills necessary to succeed in international business, or in further study. If completed successfully, your certificate from this specialisation can also be used as part of the application process for the University of London Global MBA programme, particularly for early career applicants. If you would like more information about the Global MBA, please visit www.londoninternational.ac.uk/mba This course is endorsed by CMI

À partir de la leçon

Statistical Estimation

For statistical analysis to work properly, it’s essential to have a proper sample, drawn from a population of items of interest that have measured characteristics. This week, we will cover statistical estimation, sampling distribution of the mean, point estimation, interval estimation, hypothesis testing, the Null hypothesis and look at some real life examples of their use.

Statistical Estimation - An Introduction7:27

1. Statistical Estimation0:47

2. Estimator and Estimate1:47

2.1. Point Estimator and Point Estimate1:41

2.2. Unbiased2:30

2.3. Efficiency2:25

3. Confidence Interval Estimation4:46

3.1 Confidence Intervals, Part 112:44

3.1 Confidence Intervals, Part 24:16

4. Testing Hypothesis2:40

4.1. Formulation of the Null Hypothesis and the Alternative Hypothesis3:34

4.2. Test Statistic0:56

4.2.1. The Decision Rule2:10

4.2.2. Types of Errors1:48

4.2.3. Performing the Test and the Decision Rule3:22

4.2.3.1 Hypothesis Testing: Examples7:59

5. Regression Model5:07

Statistical Estimation - Summary3:15

Rencontrer les enseignants

George Kapetanios
Professor of Finance and Econometrics
King's College London

[MUSIC]

The regression analysis is one of the most useful tools

while dealing with financial and economic models.

The regression analysis describes and evaluates the relationship

between a given variable and one or more other variables.

The final goal is to explain movements in a variable

y due to movements in one or more other variables.

xs such that x1, x2,

x3 up to xk which cause changes

in some other variables y.

We need to be aware that regressions does not mean correlation

between two variables.

The correlation measures the degree of linear association between variables.

This means that y and x are treated in a symmetrical way.

In case of regression, instead the dependent variable y and

the independent variable x are treated very differently.

The y variable is assumed to be random or stochastic.

And then, it has a probability distribution.

The x variables are however assumed to have fixed,

that is, non-stochastic values even in repeated samples.

Suppose that, based on previous studies,

there should be a relationship between two variables, y and x.

And that this theory suggest that,

an increase in x will lead to an increase in y.

If this is the case then the relationship between them can be described

by a straight line, which is the line which fit our data.

Therefore, it is of our interest determining

to what extent this relationship can be described

by an equation that can be estimated using a certain procedure.

We use the general equation for a straight line in order to

get the line that best fits the data,

such that y equal to alpha plus beta x.

Basically, we are aimed at finding the values of the parameters or

coefficients with alpha and beta.

Which would place the line as close as possible

to all of the data points taken together.

However, this equation is not complete.

Even if we add more explanatory variables in order to explain

better the behavior of y, there will always

be some omitted determinants of y from the model.

This is, for example,

because some determinants are not observable or not measurable.

There will be always what we call an error term or the random disturbance term.

Therefore, our equation becomes yt = alpha + Beta xt + ut,

where alpha is the constant term, and

beta is the slope of our straight line.

Adding the error term makes our model more realistic.

The errors represent the vertical distance between the estimate line and the data.

Therefore, our goal is to find the straight line which best fits

the data, given that a random error does exist.

The best way is to estimate the values of alpha and

beta which minimize this vertical distance.

[MUSIC]

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2018 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play