Stock price forecasting has long been an area of active research in both statistics and financial mathematics. A large portion of that work has relied on classical time series models, such as ARIMA, GARCH, and exponential smoothing, which are well understood and continue to serve as strong baselines. More recently, advances in natural language processing have introduced a new direction: using transformer-based models to extract sentiment from financial news and incorporate those signals into forecasting pipelines. This thesis sits at the intersection of both traditions. We build directly on the work of Berninger, who applied classical time series models to forecast Apple Inc.'s monthly opening stock price and found that a consensus average of all model forecasts produced the lowest root mean square error (RMSE) over a 12-month test window. We extend that framework by using Python instead of R, use different stocks, which are Tesla (TSLA), NVIDIA (NVDA), and Microsoft (MSFT), alongside Apple (AAPL) and the S&P 500, and update the sample period to span June 2010 through December 2024. Beyond this, the primary contribution of this thesis is the implementation of a transformer-based sentiment model, FinBERT, into the forecasting pipeline. Monthly price data is sourced from Yahoo Finance via the yfinance Python library. We compare seven models in total: ARIMA, ARIMA+GARCH, Holt-Winters exponential smoothing, Vector Autoregression (VAR), linear regression on ARIMA residuals, a consensus average, and a FinBERT sentiment-augmented ARIMA. Results show that ARIMA and Holt-Winters remain the strongest short-term forecasters, while FinBERT sentiment adjustments provide additional interpretable signal when tuned appropriately. The consensus average continues to perform well over the full 12-month horizon, consistent with Berninger's original finding. These results suggest that classical methods remain competitive and that sentiment-based adjustments from modern NLP models can complement but not yet replace them on monthly data.