Paradigms in Statistical Inference for Finite Populations; Up to the 1950s

Modern sample surveys started to spread after statistician at the U.S. Bureau of the Census in the 1940s had developed a sampling design for the Current Population Survey (CPS). A significant factor was also that digital computers became available for statisticians. In the beginning of 1950s, the theory was documented in textbooks on survey sampling. This thesis is about the development of the statistical inference for sample surveys. For the first time the idea of statistical inference was enunciated by a French scientist, P. S. Laplace. In 1781, he published a plan for a partial investigation in which he determined the sample size needed to reach the desired accuracy in estimation. The plan was based on Laplace s Principle of Inverse Probability and on his derivation of the Central Limit Theorem. They were published in a memoir in 1774 which is one of the origins of statistical inference. Laplace s inference model was based on Bernoulli trials and binominal probabilities. He assumed that populations were changing constantly. It was depicted by assuming a priori distributions for parameters. Laplace s inference model dominated statistical thinking for a century. Sample selection in Laplace s investigations was purposive. In 1894 in the International Statistical Institute meeting, Norwegian Anders Kiaer presented the idea of the Representative Method to draw samples. Its idea was that the sample would be a miniature of the population. It is still prevailing. The virtues of random sampling were known but practical problems of sample selection and data collection hindered its use. Arhtur Bowley realized the potentials of Kiaer s method and in the beginning of the 20th century carried out several surveys in the UK. He also developed the theory of statistical inference for finite populations. It was based on Laplace s inference model. R. A. Fisher contributions in the 1920 s constitute a watershed in the statistical science He revolutionized the theory of statistics. In addition, he introduced a new statistical inference model which is still the prevailing paradigm. The essential idea is to draw repeatedly samples from the same population and the assumption that population parameters are constants. Fisher s theory did not include a priori probabilities. Jerzy Neyman adopted Fisher s inference model and applied it to finite populations with the difference that Neyman s inference model does not include any assumptions of the distributions of the study variables. Applying Fisher s fiducial argument he developed the theory for confidence intervals. Neyman s last contribution to survey sampling presented a theory for double sampling. This gave the central idea for statisticians at the U.S. Census Bureau to develop the complex survey design for the CPS. Important criterion was to have a method in which the costs of data collection were acceptable, and which provided approximately equal interviewer workloads, besides sufficient accuracy in estimation.

Categorical Meteorological Products: Evaluation and Analysis

In meteorology, observations and forecasts of a wide range of phenomena for example, snow, clouds, hail, fog, and tornados can be categorical, that is, they can only have discrete values (e.g., "snow" and "no snow"). Concentrating on satellite-based snow and cloud analyses, this thesis explores methods that have been developed for evaluation of categorical products and analyses. Different algorithms for satellite products generate different results; sometimes the differences are subtle, sometimes all too visible. In addition to differences between algorithms, the satellite products are influenced by physical processes and conditions, such as diurnal and seasonal variation in solar radiation, topography, and land use. The analysis of satellite-based snow cover analyses from NOAA, NASA, and EUMETSAT, and snow analyses for numerical weather prediction models from FMI and ECMWF was complicated by the fact that we did not have the true knowledge of snow extent, and we were forced simply to measure the agreement between different products. The Sammon mapping, a multidimensional scaling method, was then used to visualize the differences between different products. The trustworthiness of the results for cloud analyses [EUMETSAT Meteorological Products Extraction Facility cloud mask (MPEF), together with the Nowcasting Satellite Application Facility (SAFNWC) cloud masks provided by Météo-France (SAFNWC/MSG) and the Swedish Meteorological and Hydrological Institute (SAFNWC/PPS)] compared with ceilometers of the Helsinki Testbed was estimated by constructing confidence intervals (CIs). Bootstrapping, a statistical resampling method, was used to construct CIs, especially in the presence of spatial and temporal correlation. The reference data for validation are constantly in short supply. In general, the needs of a particular project drive the requirements for evaluation, for example, for the accuracy and the timeliness of the particular data and methods. In this vein, we discuss tentatively how data provided by general public, e.g., photos shared on the Internet photo-sharing service Flickr, can be used as a new source for validation. Results show that they are of reasonable quality and their use for case studies can be warmly recommended. Last, the use of cluster analysis on meteorological in-situ measurements was explored. The Autoclass algorithm was used to construct compact representations of synoptic conditions of fog at Finnish airports.