Plate Load Test (PLT) is a significant in situ test used in soil investigation to assess ultimate bearing capacity, settlement characteristics, secant modulus, and sub-grade reaction. The test is expensive and needs a skilled workforce to conduct. Researchers are attempting to predict the outcomes mentioned from the plate load test by developing analytical models considering index properties of in situ soils. In the literature, a genetic programming and simulated annealing (GP/SA) algorithm-based model that can predict the outcomes of plate load test using the index properties such as moisture content, bulk unit weight, dry unit weight, Atterberg limits and parameters that can be obtained from grain size distribution curve was developed by researchers. However, the model is very complex in terms of the mathematical equations developed to predict the plate load test parameters. The current research mainly works on the same data and focuses on developing a model to predict secant modulus by making much simpler and fundamental machine learning regression technique, namely, multivariate regression algorithm. Various models are developed by making using of selective independent parameters and their model performance was studied. The developed models were also studied for the effect of interaction parameters. All the multivariate regression models are developed in Microsoft Excel software.