Vehicles are a major source of transportation greenhouse gas emissions and the need to accurately quantify and monitor transportation-related emissions from vehicles is nowadays essential. Vehicle emissions are complex functions to be approximated in practice due to many variables affecting their outcome. The aim of this research is to study factors affecting different types of vehicle emissions on Egyptian roads. Models were calibrated using vehicle emissions records collected in the period 2018/2019 and data were recorded in the field for eight types of vehicles. Emission data were classified into three categories according to the fuel type (Diesel, Natural Gas, and Petrol Vehicles). A comparative analysis of various statistical modelling techniques was used to predict vehicle emission rates as a function of six independent variables for vehicle emissions. The Linear Regression Model with Link Function of a Log was found to be the best generalized regression model to represent the correlation between CO2, CO and NOX emissions for Diesel vehicles, whereas the Linear Regression Model with Link Function of Identity was a good representative for the relationship of HC emission for Diesel vehicles. Natural Gas and Petrol vehicle emissions (CO2, CO, HC, and NOX) were best represented with the Linear Regression Model with Link Function of Log. Amongst the studied independent variables, changes in the ambient pressure (P) and numbers of rotations per minute for vehicle engine (RPM) were found to be directly proportional with gas emission for all the three types of vehicles in this study. In addition to these factors, increase of emissions from Diesel vehicles was also related to increasing vehicle speed (V), ambient temperature (T) and relative humidity (RH), whereas emissions from Natural Gas and Petrol vehicles were found to increase also with road grade (G) (both), and ambient temperature (T) (Natural Gas only).