A more detailed description will be given of the process for obtaining the prediction values of the spectral reflectances (the process in Step S160). At the point of time when the process of Step S160 is started, the quintic equation, for example, such as the above Equation (4) is obtained as the second relational equation for each of the prediction target gradation values for the reference color. Moreover, as mentioned above, the data of the spectral reflectances is composed of 36 reflectances. Accordingly, in Step S160, for each prediction target gradation value, the 36 reflectances which are the data of the spectral reflectances of the maximum gradation values for the prediction target color are assigned one by one to the corresponding second relational equation (the relational equation which represents the relationship between the spectral reflectances of the maximum gradation value and the spectral reflectances of the relevant prediction target gradation value), whereby the 36 reflectances which serve as the data of the spectral reflectances of the relevant prediction target gradation value for the prediction target color are obtained. In the present embodiment, when the second relational equation is calculated, normalization is performed so that the spectral reflectances of the minimum gradation value are 1. Hence, the 36 reflectances obtained from the second relational equation are subjected to denormalization based on the actual spectral reflectances of the minimum gradation value. The denormalization is a process for returning normalized data to unnormalized data.