This thesis studies the effect of Tariffs and Non Tariff Measures on Imports, by using bilateral trade data. The data is specified on 4-digit HS-code, with a total of 1223 different products. Theory describes that using OLS for the estimation of the gravity equation has three major shortcomings: the bias created by the logarithmic transformation of the dependent variable, the violation of the homoskedasticity assumption, and the zero trade values who are removed from the analysis or are manipulated by adding a small number. Theory suggests to use Poisson Models to solve these problems. In the analysis we compare OLS, Poisson Model, Negative Binomial Poisson Model, and Zero Inflated Negative Binomial Poisson Model and apply them to the gravity equation. As control variables, common border and common language are used. The main conclusion is that tariffs have a robust significant negative impact on imports over all the different econometric specifications, and increasing tariffs raise the probability of a zero trade flow. There is evidence that NTMs have a positive impact on imports and the probability to trade. The effect of NTMs on imports is an overall effect. Individual effects of several types of NTMs are unknown and can still be negative. The effect found possibly suffers from several problems, such as the underlying definition and reverse causality. Robustness checks show that it is likely that at least the probability to trade is negative related to having NTMs in place