8/19/2019 GDB-2-2.docx

1/1

2nd GDB of MGT-601

Q: Distinguish between Statistical Estiation and !"#othesis Testing$

Both estimation and testing are concerned with a parameter, which should be a meaningfulquantity. In statistics estimation refers to the process by which one makes inferences about a

 population, based on information obtained from a sample.

Statisticians use sample statistics to estimate population parameters. For example, sample means

are used to estimate population means; sample proportions, to estimate population

 proportions.n estimate of a population parameter may be expressed in two ways!

Point estimate. point estimate of a population parameter is a single "alue of a statistic.

For example, the sample mean x is a point estimate of the population mean #. Similarly,

the sample proportion  p is a point estimate of the population proportion  P .

Interval estimate. n inter"al estimate is defined by two numbers, between which a

 population parameter is said to lie. For example, a $ x $ b is an inter"al estimate of the

 population mean #. It indicates that the population mean is greater than a but less than b.

statistical hypothesis test is a method of statistical inference. %ommonly, two statistical data

sets are compared, or a data set obtained by sampling is compared against a synthetic data set

from an ideali&ed model. hypothesis is proposed for the statistical relationship between the two

data sets, and this is compared as an alternati"e to an ideali&ed null hypothesis that proposes norelationship between two data sets. 'he comparison is deemed statistically significant if the

relationship between the data sets would be an unlikely reali&ation of the null hypothesis

according to a threshold probability(the significance le"el. )ypothesis tests are used in

determining what outcomes of a study would lead to a re*ection of the null hypothesis for a pre+

specified le"el of significance. 'he process of distinguishing between the null hypothesis and the

alternati"e hypothesis is aided by identifying two conceptual types of errors type- type/0 and

 by specifying parametric limits on e.g. how much type - error will be permitted.

Statistical hypothesis tests define a procedure that controls fixes0 the probability of incorrectly

deciding  that a default position null hypothesis0 is incorrect. 'he procedure is based on howlikely it would be for a set of obser"ations to occur if the null hypothesis were true. 1ote that this

 probability of making an incorrect decision is not  the probability that the null hypothesis is true,

nor whether any specific alternati"e hypothesis is true. 'his contrasts with other possible

techniques of decision theory in which the null and alternati"e hypothesis are treated on a more

equal basis.