本研究探討分析師預測是否可以領先油價，據以形成新油價預測模式。過去油價預測主要賴以供需模型、時間序列、期貨價格、總體變數以及機器學習模型來預測原油價格。而本研究以分析師營收預測來預測油價，因油價漲跌會反映在原油產業的營業收入，故營收預測會領先於油價，本研究主要以營收預測來對未來油價進行解釋，並以盈餘預測作為附加檢測，並進一步探討發生特殊事件或結構性改變，分析師對油價的預測能力是否會因此改變。研究樣本為2007年至2018年4月北美原油上游產業，以西德州原油價格作為研究標的，來檢驗分析師預測對油價是否具解釋能力。實證結果顯示，分析師營收及盈餘預測成長率對油價並無解釋能力，而分析師營收及盈餘預測修正與未來油價呈現正向關係，在金融海嘯期間，分析師預測能力顯著下降；但以頁岩油為例之技術進步，分析師預測能力顯著上升。;This study discusses whether analysts can predict crude oil prices. In the previous studies, oil price forecasts were based on the supply and demand model of crude oil price. In other words, it means forecasting crude oil price fluctuations based on crude oil inventory levels. In this study, revenue is used to forecast oil prices, and the revenue of the petroleum industry is a reflection of the oil price. Therefore, this study will use revenue forecast to carry out future crude oil prices prediction and use earnings forecast as an additional test. Furthermore, we also detect the relationship between revenue forecast and oil price when specific exogenous events or structural change is happening. North American crude oil upstream firms from 2007 to 2018 and West Texas Crude Oil Price (WTI) was taken as the research samples to test whether revenue forecast can predict oil prices. Empirical result shows that there is no relationship between analysts’ revenue & earnings forecast growth rate and future oil price. But, a positively significant relationship between analysts′ revenue & earnings forecast revision and future oil price. In the period of financial tsunami, the analysts’ forecast ability decreases. But, analysts’ forecast ability increases in technology improvement of shale oil extraction.
