دانلود رایگان مقاله مدل سازی بازار نفت: تحلیل مقایسه روش عامل اساسی و پنهان

Abstract

We formally compare fundamental factor and latent factor approaches to oil price modelling. Fundamental modelling has a long history in seeking to understand oil price movements, while latent factor modelling has a more recent and limited history, but has gained popularity in other financial markets. The two approaches, though competing, have not formally been compared as to effectiveness. For a range of short- medium- and long-dated WTI oil futures we test a recently proposed five-factor fundamental model and a Principal Component Analysis latent factor model. Our findings demonstrate that there is no discernible difference between the two techniques in a dynamic setting. We conclude that this infers some advantages in adopting the latent factor approach due to the difficulty in determining a well specified fundamental model.

4. Conclusion

In this study, we conduct a formal model comparison of fundamental factor and latent factor modelling approaches to capturing oil market movements. To the author's knowledge such a formal comparison has not been conducted to date and so our study fills this gap. We additionally contribute by means of conducting our model comparison exercise across the term structure of oil prices to examine whether there is a difference in model fit based on contract maturity. Prior literature (Dempster et al., 2012) suggests that different factors impact on the short-, medium- and long-term, and so our analysis across the forward curve provides insights into the ability of the respective fundamental and latent factor models to capture these term structure movements. In a final contribution, we use our model comparison setting to formally test whether speculation is a fundamental factor affecting oil prices. We find the fundamental factor model and latent factor model specifications to be of equal fit to the overall population of oil futures prices. This finding is significant given the fundamentally different approaches taken to modelling oil prices. Latent factor models, which have received much less attention in the literature, outside of derivatives applications, offer real merit it seems relative to the popular approach of fundamental modelling. In particular, latent factor models avoid the need to search out and identify appropriate economic factors, in so doing avoid criticisms about omitted variable and/or irrelevant variable bias, and are accessible and easily implemented being purely driven by the data of interest. In principle, the principal components should contain all of the information contained in a well specified fundamental factor model. The novelty of our study is the formal model comparison of fundamental factor and latent factor approaches. The work should motivate further research to appraise the two alternative strands of modelling, using an extended range of fundamental and latent factors. This should be performed on a more expansive cross-commodity basis to see if the findings observed here hold across markets.