The usual econometric model for annual trends is expanded with values of x. From the logarithms of the formulas is derived a linear model that permits estimation of coefficients through regression analysis. The years 1965 to 1975 are used as a base period. The model is applied to a number of offenses and offense categories using multiple stepwise regression. Calculations are done with the SPSSH package of the IBM 370/158 computer of the Central Computer Institute in Leiden. Using this system a good analytic connection can be made between the time and the prevalence of offenses. The model proves especially useful in specifying trends. Most models estimate trends by comparing the extent of crime in the year N + 1 with crime in the base year N. Increases and decreases are thus fixed to an index year some years back. A more realistic picture can be obtained by relating figures on crime extent to the figures over a period of past years, as the model described does. Thus, the increase in criminal offenses in 1976 under the Dutch criminal code is 15 percent over the base year 1975 but only 7.8 percent over the previous 10-year period. While aggressive offenses in 1976 increase 17.4 percent over 1975, they are 3.4 percent lower when compared to the prior 10-year period. By comparing the figures calculated in the two different fashions, a more precise image of long term trends is obtained. In this way more sound conclusions about the course of crime in the Netherlands can be reached. Tables are supplied.