Impact analysis methods are commonly employed to reduce the likelihood of encountering faulty or unexpected behavior from a software program as a result of developers' oblivious modifications. In this paper, we propose a static impact analysis technique that creates clusters of closely associated software program files based on their co-modification history in the software repository. The proposed method benefits from dimensionality reduction techniques to reduce the complexity of the collected information and perform the impact analysis process faster. The method has been tested on four different open source project repositories, namely Firefox, Firebird, Thunderbird, and FileZilla. The results of the impact analysis method performance in terms of precision (impact set identification accuracy) and execution time cost have been reported in this paper. The proposed method shows promising behavior when used with several specific clustering techniques such as DBscan and X-Means.