14
W. K. Hardle, R. A. Moro, and D. Schafer
AR (SVM K5+K29+K7+K33+K18+K21+K24+K*)
Variable No.
Fig. 10. Accuracy ratios for SVM models with eight variables. Each model includes
variables K5, K29, K7, K33, K18, K21, K24 and one of the remaining. Box-plots are
estimated basing on 100 random subsamples.
6 Colour Coding of PDs
The RGB colour space is based on three primary colours, red, green and blue,
that are mixed to produce other ones. It is the colour coding scheme that is
used in monitors and TV. It is, however, inconvenient for colour coding since
we would like to make adjustments only to the channel responsible for colour
while keeping lightness and saturation constant. This can be achieved with
the HLS colour space.
We will represent the probability of default (PD) estimated with the SVM
as two-dimensional plots where each colour represents a specific PD. The PD
is a number that can be represented on a gray scale, e.g. in the RGB encoding
as (i, i, i) with i changing from 0 to 255, e.g. the colour R=255, G=0, B=0
corresponds to red, and R=255, G=0, B=255 to violet, etc..
The HLS stands for hue, lightness (or luminance) and saturation. By ad-
justing only hues and keeping luminance and saturation fixed we can generate
simulated colours from the range shown in Figure 14. The pure red colour
corresponds to H=0 or 360, the pure green to H =120 and the pure blue to
H=240.
More intriguing information
1. Unemployment in an Interdependent World2. Crime as a Social Cost of Poverty and Inequality: A Review Focusing on Developing Countries
3. International Financial Integration*
4. Examining the Regional Aspect of Foreign Direct Investment to Developing Countries
5. Effects of red light and loud noise on the rate at which monkeys sample the sensory environment
6. AMINO ACIDS SEQUENCE ANALYSIS ON COLLAGEN
7. Group cooperation, inclusion and disaffected pupils: some responses to informal learning in the music classroom
8. Population ageing, taxation, pensions and health costs, CHERE Working Paper 2007/10
9. CURRENT CHALLENGES FOR AGRICULTURAL POLICY
10. The name is absent