Department of Mathematics and Statistics

Pahkinen, Erkki, Emeritus Professor

Subject:
statistics
Room:
MaD366

Research Interests:
Survey Sampling Methods

Statistics as a field of science has developed and continues developing new methods to ensure the controlled collection and analysis of research data. The primary source of the data is normally a database, the administration of which is responsible for its quality. The sources for the data and databases can come from designed statistical sampling methods as well as experimental methods applied to diverse situations.

A straw poll carried out in our country is an example of a survey research. The 4.2 million citizens at voting age are the basis of the straw poll. The knowledge of statistical sampling theory is required when a representative random sample of 2000 voters is selected into a telephone interview. Because the research data is a probability sample, the computational results include randomness, and thus the computed support percentage is an approximation or estimate. For this reason, the margin of error and corresponding estimates for the results are reported. For example, the true value of the support percentage among all voters (population) has a ±2 percent margin of error around the estimate.

The conversion of statistics into graphs makes interpretation essentially easier for the user of the statistical information. However for example, a monthly measured time-series contains so much vibration that the curves appear as broken lines. It is possible to forecast trends by using statistical methods and to present the important or significant findings in a format that is visually attractive, interesting, and easier to digest.

Anybody or any organization – a citizen, an enterprise, media and official administration,may in certain situations need statistical information concerning their operating environment. The application of statistics is essential in the research institutes of different fields. These institutes produce and process many types of research data. The work of statisticians is required to help interpret the significance of the data produced.

Statistics as a field of science has developed and continues developing new methods to ensure the controlled collection and analysis of research data. The primary source of the data is normally a database, the administration of which is responsible for its quality. The sources for the data and databases can come from designed statistical sampling methods as well as experimental methods applied to diverse situations.

A straw poll carried out in our country is an example of a survey research. The 4.2 million citizens at voting age are the basis of the straw poll. The knowledge of statistical sampling theory is required when a representative random sample of 2000 voters is selected into a telephone interview. Because the research data is a probability sample, the computational results include randomness, and thus the computed support percentage is an approximation or estimate. For this reason, the margin of error and corresponding estimates for the results are reported. For example, the true value of the support percentage among all voters (population) has a ±2 percent margin of error around the estimate.

The conversion of statistics into graphs makes interpretation essentially easier for the user of the statistical information. However for example, a monthly measured time-series contains so much vibration that the curves appear as broken lines. It is possible to forecast trends by using statistical methods and to present the important or significant findings in a format that is visually attractive, interesting, and easier to digest. Anybody or any organization – a citizen, an enterprise, media and official administration,may in certain situations need statistical information concerning their operating environment. The application of statistics is essential in the research institutes of different fields. These institutes produce and process many types of research data. The work of statisticians is required to help interpret the significance of the data produced.

Publications in TUTKA system

Homepage:
http://users.jyu.fi/~pahkinen/