model selection and shows some applications of the methods. Especially,

plant ecological and worm ethological data with spatial information are

treated. Characteristic problems arising in each data are shown through the

data analysis of explanatory variables in base models.

Firstly, the problem of selecting topographical attribute as the

explanatory variable is discussed through contingency table statistics. As an

example, an analysis of observed data about relative altitudes and about

distribution of moss is shown. In order to consider the relationship between

topography and distribution of plants, data of altitude and mosses in study

plots in continental Antarctica were collected by the present author when he

had a chance to visit there. The altitude data was processed as an

explanatory variable of model about the existence of moss. As one of the

important topographical attributes that are calculated from altitude data,

the relief from standard surface was adopted. Under the assumption that the

probability of moss existence is proportional to the value of residuals from

standard surface, the strength of relationship between topography and moss

distribution was obtained by using 2 X 2 contingency table statistics. It

suggested that a simpler standard surface had a better ability, than an

adjusted standard surface, of determining topographical attribute which

strongly related to the moss existence. Then, the standard surface estimated

by robust methods presented in this paper had a little better ability than by

a prevailing least-square method. Totally, the accurate regression methods

were overfitting as a method of determining the standard surface.

Secondly, a modeling with spatial information is treated. The problem

of selecting a model is discussed through linear models and linear mixed

models. As examples, the analyses of the distribution of dwarf pine in mixed

forest and of the moving track of nematode are shown. In the analysis of

dwarf pine, data of forest study was collected in central Kamchatka by the

research group in which the present author has participated. There, the size

of stem diameter at the base of the lowest live branch of dwarf pine

individuals and their distance from canopy trees were used for parameters of

growth models. Consequently, as in many previous works of growth model

about canopy trees, an adequateness of using the inverse of distance in

competitive effect terms was shown. In the analysis of nematode, data

collected by an automatic tracking system of the center of gravity about the

individuals at an experiment of learning action to detestable odorant for

nematode was used. The avoidance action of nematode from the odorant was

explained by the factors involving time and coordinate. In the detestable

odorant (2-nonanone) exposed condition, the selected linear model retained

the term of distance with positive coefficient and did not retain the term of

time. On the other hand, the selected linear mixed models retained the terms

of time with positive coefficient and the distance with negative coefficient in

selected models about avoidance action. However, in the solvent (ethyl

alcohol) exposed condition, different variable selection rules lead quite

different results in linear mixed models and stable interpretation cannot be

done. As a result of model selection, it is shown that the linear mixed models

have substantially better reproduction ability of moving track than the other

models.

In summary, by applying statistical methods and models presented in

this thesis, it is shown that many significant results in ecology and ethology

are derived which are useful and suggestive in respective scientific fields., 総研大甲第1198号}, title = {Statistical analysis of plant ecological and worm ethological data - Some viewpoints of explanatory variables in base models -}, year = {} }