Module 5 Estimation of Parameters
Problems concerning the estimation of parametrical values are basic to
educators, scientists, engineers, and businessmen. A n error in estimation would
lead to biased results, and as a consequence, biased decisions as well. Thus, the
study of the estimation of parameters plays an important role in any area of
discipline. Let us say a manager is researching on the problems regarding the daily
outputs of his 2,000 workers. It would be too costly and time-consuming to study
the entire population of 2,000. Instead, focusing on a sample of 50 to 100
selected employees to determine their average output greatly simplifies the
problem. This called a point estimate. The study of the output of a group of
workers by intervals is an example of an interval estimate of a parameter.
Proper estimation of population parameters results in accurate data and
quick solutions to real-life problems in any field of study.
Key Words :
Estimation – the process of making inferences about a population based
on information obtained from the sample.
Point estimate – the sample mean x̄ of the population or mean µ . It is the
numerical value which gives an estimate of a parameter.
Interval estimate – a range of values used to estimate the parameter. It can
be calculated using two numbers or values which may or may
not contain the value of the parameter being estimated.
Statistics – measurement or observations from a sample population. This is
usually denoted by Arabic letters (e.g. x.̄ = average, sd = standard
deviation, p=proportion)
Parameters – measurement from the universe or target population. This is
usually unknown and differentiated from statistics by Greek
Letters (e.g. µ=population mean, σ = standard deviation,
π=proportion)
Two Types of Making Inferences About the Population
1. Estimation
Estimation uses a sample statistic to approximate the unknown
population parameter. A point estimate is the best estimate, in some
sense, of the parameter based on a sample. It should be obvious that any
point estimate is not absolutely accurate. It is an estimate based on only a
single random sample. If repeated random samples were taken from the
population, the point estimate would be expected to vary from sample to
sample. Interval estimate on the other hand consists of two numbers, an
upper and a lower limit – these are used to serve as the bounding values
within which the population parameter is expected to lie with a certain
degree of confidence. Interval estimation is important in statistical
inference, as it important in statistical inference, as it takes into account
the sampling error and helps in making the results more precise.
Confidence Intervals
General Formula:
95% Confidence Interval = Point estimate ±
(Reliability coefficient x Standard error )
Reliability Coefficients – These are taken from the table of
probabilities (e.g. T-table, Z-table, chi square table). They are dependent on
the given confidence level. As confidence level increases, the reliability
coefficient also increases.
Confidence Reliability Coefficient (Z
Level value)
90% 1.64
95% 1.96
99% 2.58
Point Estimates and Standard Errors
Standard errors are a function of the measure of variation and
sample size
Point Estimate Standard Error
Sample proportion Std. error =
(p)
Sample mean ('x̄' ) Std error
Where : p = proportion; n = sample size; sd=standard deviation
