dHourly wind speed data are simulated at few sites for depicting extreme values during various return periods by using transition matrix (TM) methods in northwestern region of Turkey. The simulation is performed by first order Markov chain model. The wind speed time series is divided into states depending on the arithmetic average and the standard deviation. Wind speed observations are used to generate series of the same length real time series and then the statistical parameters in addition to the autocorrelation functions as well as the maximum and minimum values are preserved in the synthetic sequences. Once the control and validation of the model is confirmed, it is then used for generating synthetic series of various lengths of any desired duration. The maximum wind speed from each 1000 run is extracted, and their mean value is taken as the extreme wind speed for a return period equal to the length of simulation. It is observed that for short return periods, in other words, for high risk levels the results obtained from the synthetic time series are close to the conventional risk calculations by conventional extreme value analysis whereas for long durations conventional approaches overestimate the wind speeds at any given risk level. It is concluded that the results from the Markov model may be more realistic since one would expect some theoretical maximum to wind speeds at any location imposed by atmospheric characteristics.