Future energy sources are expected to have a substantial contribution from renewable sources. This is expected to equal or surpass liquid fuels as the largest source of energy by 2050 [1] [2]. The ocean accounts for nearly 71% of Mother Earth’s total surface area. In combination with other renewable sources (solar, wind, etc.), harvesting energy from oceans can provide the cleanest form of energy to meet global demand. This thesis focuses on analyzing a special kind of energy harvesting device called: Oscillating Water Column (OWC). OWC works on the principle of air pressure oscillations (compression and suction/decompression) in the vertical OWC chamber induced by an oscillating water column due to the random wave actions on the exposed submerged inlet to the OWC structure. The back-and-forth movement of air (due to compression and suction) turns a turbine that rotates the rotor connected by a gear system to the generator. The overall process eventually generates electricity that can be stored and distributed by a grid system. It is noted that irrespective of the direction of the air, the turbine always turns in the same direction due to the design of the blades. These are termed self-rectifying turbines. In the thesis, first, an exhaustive literature survey of the OWC concepts and applications is done. Next, the equations of motion (EOM) for the OWC are formulated based on the unsteady Bernoulli equation and the wave pressure based on the linear potential wave theory. This yields a set of two coupled nonlinear integro-differential equations that describe the OWC system. This set of nonlinear equations does not lend itself to an exact analytical solution. A thorough stochastic dynamics approach using two independent methods is employed to treat these equations. The first method uses the Monte Carlo technique in the time domain. The latter employs a quite comprehensive statistical linearization technique working in the frequency domain. To ensure that the results from the Monte Carlo assessment are reliable, a comparison of the water column height inside the OWC chamber, obtained from the Monte Carlo assessment, is made with that obtained from a pertinent laboratory test. Next, an equivalent linear system that resembles the nonlinear system in a statistical sense is sought to estimate the response using the statistical linearization technique. This set of equations representing the equivalent linear system is then solved iteratively to estimate the dynamical response statistics of the system. Next, the reliability of the solution procedure is assessed by comparing relevant Monte Carlo data versus results obtained from the statistical linearization technique. To evaluate the performance of the statistical linearization technique, requisite computations are done using various random input wave characteristics relating to the JONSWAP ocean waves spectrum. Several parametric studies are also conducted.