Silicon photonics has significantly matured in the near-infrared (telecommunication) wavelength range with several commercial products already in the market. More recently, the technology has been extended into the mid-infrared (mid-IR) regime with potential applications in biochemical sensing, tissue photoablation, environmental monitoring and freespace communications. The key advantage of silicon in the mid-IR, as compared with near-IR, is the absence of twophoton absorption (TPA) and free-carrier absorption (FCA). The absence of these nonlinear losses would potentially lead to high-performance nonlinear devices based on Raman and Kerr effects. Also, with the absence of TPA and FCA, the coupled-wave equations that are usually numerically solved to model these nonlinear devices lend themselves to analytical solutions in the mid-IR. In this paper, an analytical model for mid-IR silicon Raman lasers is developed. The validity of the model is confirmed by comparing it with numerical solutions of the coupled-wave equations. The developed model can be used as a versatile and efficient tool for analysis, design and optimization of mid-IR silicon Raman lasers, or to find good initial guesses for numerical methods. The effects of cavity parameters, such as cavity length and facet reflectivities, on the lasing threshold and input-output characteristics of the Raman laser are studied. For instance, for a propagation loss of 0.5 dB/cm, conversion efficiencies as high as 56% is predicted. The predicted optimum cavity (waveguide) length at 2.0 dB/cm propagation loss is ~ 3.4 mm. The results of this study predict strong prospects for mid-IR silicon Raman lasers for the mentioned applications.