Urban transportation planning heavily relies on travel forecasting. Travel forecasting models determine whether an increase in road capacity is required, if transit services need to be changed, and how land use patterns need to be adjusted. Travel and activities serve various purposes to satisfy people's needs. It's true that every person's travel is influenced by constraints, which may include changes in spatial, temporal, demographic, socioeconomic characteristics, etc. Technological advancements such as smartphones, social changes like 24-h stores, immigration movements, etc., are making travel patterns unpredictable throughout the entire seven-week period. Therefore, in-depth studies are recommended to capture people's travel behavior over time. The 2017 National Household Travel Survey (NHTS) summary statistics for demographic characteristics and travel data were utilized to estimate a Negative Binomial regression model for household trip-based generation in Michigan. This study evaluated several factors affecting daily household trips based on their importance to understand their differences. Different statistical methods, including Lasso, Ridge, Elastic Net, Random Forest, and Pearson Correlation, will be used to identify and select the most relevant variables. The results indicate that household size, household income, travel day, and the number of young children in household were the most important factors affecting Michigan's travel patterns. The study findings aim to provide useful information for stakeholders involved in transportation planning and policymaking.