Hidden Markov Models are based on Markov chains and enabled to forecast of hidden states and observations with data in the past. Generally economic indicators are analyzed with some econometric applications and Hidden Markov model has limited scope for analyzing them. The aim of this study is to demonstrate the applicability of Hidden Markov models could be applied on the foreign direct investment, which has an undeniable importance on economic phenomenon and contributes savings of developing countries. It also provides improvements in the macroeconomic indicators and development of the countries. The case of Turkey in 2009:09 – 2013:12 period is used in order to identify the hidden states for hidden Markov model accurately. Then, an econometric analysis is made on the determinants of a country took place in the theory. Logarithmic - linear model is estimated as it is an appropriate model with independent variables; foreign trade volume, the unemployment rate and the CPI. After the end of the data range, with the three main problems of Hidden Markov models, foreign direct investment coming into the country (observations) and its determinants (hidden states) are estimated with the previous month values. Besides, the model parameters are optimized for the desired sequence of the observation. Estimated results are compared with the actual values and they are interpreted. Under the case of having a big data and identified hidden states correctly, hidden Markov models can be said to give accurate results for longer periods.