A new method using empirical likelihood is introduced to detect changes in coefficients of a high-dimensional linear model. The goal is to test if there is a change in the regression coefficients. The method simplifies the testing process and shows that the test statistic follows a normal distribution under the null hypothesis. This is different from the chi-square distribution used in models with a fixed number of variables. When there is a change in coefficients, the test statistic diverges, allowing for the calculation of confidence intervals for the differences in parameters between the two phases. Simulation studies confirm the effectiveness of the proposed method.