Correlation, Method of Least Squares, Linear Regression and the General Linear Model - Geomathematics: Theoretical Foundations, Applications and Future Developments - page 128

Fig. 4.9 3-D representation of three copper probability maps for same area as depicted in Fig. 4.7 .
The main peaks near Val d'Or, Noranda and Timmins fall approximately on east-west line
pointing upwards to the left in these diagrams. (a) Probability index (based on 27 control cells
shown in Fig. 4.5 ), scaled with respect to control area of 50 cells shown in Fig. 4.5 ; result is based
on stepwise regression with Q ¼ 0.5; 26 variables were included. (b) Ditto, as computed from
19 control cells east of 600 km E line of Fig. 4.7 ; control area consisted of the 25 cells near
Noranda only. (c) Ditto, from 8 control cells in western part of area; control area of 25 cells near
Timmins only. Note that b resembles a, but c underestimates a in the eastern part of the region
(Source: Agterberg 1974 , Fig. 117)
about three times as much as was contained in the 41 copper deposits in 1968. The
41 copper deposits in the 1968 dataset occurred in 27 “copper cells” measuring
10 km on a side and belonging to the original set of 644 cells for which copper
potential was estimated in 1971. Figure 4.10 shows most copper cells, with
(A) locations of 1968 copper cells, (B) locations of 1977 copper cells including
two new cells with new discoveries, and (C) locations of copper cells in 2008 each