As general conclusions of the performed regression analysis it can be established that the increase of both agricultural and forest land areas increases the sales price less than proportionally (elasticity < 1) and that the area of arable land, the forest land and the dummy variables explain, in 1961, about 43 %, in 1962 29 % and in 1966 about 39 % of the changes of the sales prices of farm estates in the whole country. From the provincial models it can be observed that the variance of sales prices explained is highest in the field cultivation territory, i.e. the provinces Turun ja Porin lääni, Hämeen lääni and Kymen lääni, in which it varies from 42—58 %. On the other hand, in the ’’forest region” of Finland, i.e. in the provinces Keski-Suomen lääni and Pohjois-Karjalan lääni, the variance explained was also high, being 33—52 %. The lowest R2‘s were found in the provinces Mikkelin lääni, Vaasan lääni and Oulun lääni. Thus many other factors, such as the quality of the land, the location of the estate with regard to population centers, the buildings, roads and environmental factors, which could not be included in this research, are evidently important arguments of the sales price. It would thus be necessary to expand the research in suchwise that the explanatory efficiency of the mentioned variables would be tested, which would, again, mean a considerable decrease of the size of the sample, because it would be necessary to obtain the data by means of expensive field research and interviews (BRIGHAM 1965). As it is difficult to find out the movements of the price level of farm estates by means of the regression analysis, here will be given the sales prices of estates in the whole country by averages, according to size categories, which averages are compared with the development of the indicators of the general price level (the basic year being the same in all series, 1961 = 100). 1961, 1962, 1966; The wholesale price index: the general index of home market goods (1935 = 100) 100 102 122; The producer price index for agricultural products (1937—39 = 100) 100 102 132; The cost of living index (X 1951 = 100) 100 104 131; The sales prices of farm estates in the whole country according to the total area: area 2—5 hectares 100 116 144; „ 5—10 „ 100 113 157; „ 10—20 „ 100 110 138; „ 20—50 „ 100 104 151; „ 50— „ 100 127 148; The sales prices of farm estates in the whole country according to the area of arable land: area 2—5 hectares 100 110 136; „ 5—10 „ 100 106 149; „ 10-20 „ 100 101 156; „ 20—50 „ 100 135 160; The sales prices of farm estates in the whole country according to the forest land area: area 2—5 hectares 100 122 154; „ 5-10 „ 100 109 156; „ 10—20 „ 100 100 140; „ 20—50 „ 100 113 163; „ 50— „ 100 125 131. As can be observed from the above table and figure 1 the sales prices of farm estates have risen considerably more in the years 1961—1966 than the general price level as measured by the Wholesale Price Index or the Cost of Living Index. Relatively the greatest price increase has taken place in farm estates with 5–10 hectares, and thereafter in sequence farm estates with 20—50 hectares, more than 50 hectares, 2—5 hectares and 10—20 hectares. It is also to be observed that the sales prices of all farm estates have risen more than the Producer Price Index for Agricultural Products.
Statistical Method to predict Agricultural CPI
