The relationship between stocks and oil prices- 股票與油價之關係 (2)
We apply Hamilton’s method, using daily data. For reasons of data availability, we start the sample in mid-2011. Data are in percentage changes (more precisely, log changes), except for the change in the ten-year rate, which is a simple difference. We estimate the equation for data through mid-2014, then use that equation to predict the oil price that would have obtained if the only shocks to the oil market had been on the demand side (the light blue line in Figure 3). The estimated coefficients on all variables (see Appendix 1 at end of post) are highly significant, both economically and statistically.
我們使用日常資料以應用漢米爾頓的方法。為了資料取得的可能性，我們在 2011年中旬開始這個範例。資料是在用百分比變化 （更確切地說，用日誌），除了十年期利率的變化，這是簡單的差異。我們估計到 2014 年中旬，資料的方程式，然後使用該方程式來預測油價，如果對石油市場唯一的一些衝擊，已經出現在需求方面 （在圖 3 中之淺藍線），就可取得。（見在本文末端之附錄 1）所有變數的估計係數，在經濟和統計上都極具意義。
Comparing the predicted and actual decline in oil prices, we find that something in the range of 40-45 percent of the decline in oil prices since June 2014 can be attributed to unexpectedly weak demand. (This range is very similar to that obtained by Hamilton in a different sample.) The results are not much affected if we drop changes in the value of the dollar from the equation, or if we replace the ten-year Treasury yield with the slope of the yield curve (the difference between the two-year Treasury yield and the ten-year yield).
比較油價預測和實際下降，我們發現，自從2014 年 6 月以來，40-45%石油價格的下降，可以歸因於出人意料之外的疲弱需求。（這個範圍非常類似於，漢密爾頓在不同的樣例中所得到的範圍）。這些結果並不太會受到影響，如果我們從方程，降低美元的價值的變化，或如果用收益率曲線的斜率取代我們十年期收益率，(兩年期國債收益率和之間的差異）。
Using this breakdown, we can also re-examine the correlations of changes in stock prices with changes in oil prices. This time we break the changes in oil prices into the part associated with demand (according to this method) and the residual portion (which presumably includes most supply factors). The correlations of stocks and demand-related changes in oil prices are shown in Figure 4, and the correlations of stocks and the residual are shown in Figure 5.
使用這種解析，我們還可以重新檢查股價與油價變化的相關性。這一次我們把把石油價格的改變裂分成與需求（按此方法）和殘餘部分（其中可能包括大多數供應因素）相關聯的這一部分。股票與需求相關的石油價格變化的關係圖 如4 所示，股票和殘差的相關性如圖5 所示。
As expected, the correlation between stock prices and the demand component of oil is higher (about 0.48, on average) than the correlation between stock prices and the oil price overall (0.39). Both of these sets of correlations in turn are higher on average than the correlations between stock prices and the residual component of the oil price (which averages about 0.16 in our sample). That’s consistent with the idea that when stock traders respond to a change in oil prices, they do so not necessarily because the oil movement is consequential in itself, but because fluctuations in oil prices serve as indicators of underlying global demand and growth.
不出所料，股票價格和石油需求組成要素之間的相關性是較高於（約平均 0.48，）比股價和整體油價(0.39)之間的關係。這兩套相關性平均都高於，股價格和油價 （在我們的示例中，平均約 0.16） 的殘餘組件之間的相關性。這是與此觀點一致，當股票交易員在石油價格變化的反應時，他們這樣做不一定因為油價波動本身是結果，而是因為油價波動指示出，基本的全球需求和成長。
On the other hand, since even the residual component of oil prices is positively correlated (on average) with stock prices movements, we have to conclude that the demand explanation (at least, given our admittedly noisy measure of demand) is not the full story.
另一方面，因為即使油價的殘留成分的與股價變動呈正面相關（平均），我們就不得不得到結論，（至少，鑒於我們喧擾的需求措施） 需求的解釋並非完整的情形。
A second possible reason for the positive stocks-oil correlation is based on the observation that recent market moves have been accompanied by elevated volatility. If investors retreat from commodities as well as stocks during periods of high uncertainty and risk aversion, then shocks to volatility may be another reason for the observed tendency of stocks and oil prices to move together. To test whether changes in risk can help explain the oil-stocks relationship, we augmented the Hamilton-style equation for oil prices with daily percentage changes in the VIX, which measures the volatility of stock indexes.[3] The VIX enters the estimated equation with the expected negative sign (oil prices tend to fall when volatility is high) and with high statistical significance (see Appendix 2 at end of post for details).
油價與股價相互關係的第二個可能原因，乃基於觀察到，近期的市場走勢，伴隨波動性的上升。 如果投資者在不確定性和厭惡投資的風險下，退卻不買大宗商品和股票，那麼對波動性的衝擊，也許就是觀察到油、股價格同步移動趨勢的另一個原因。為了要測試風險的變化，是否可以説明解釋石油、股票的關係，我們增大石油價格與措施的股票指數的波動性，用日常百分比在VIX中的變化，此VIX乃測量股票指數的波動。 [3] vix 用預期的負號指數，進入估計的方程式，(當波動率高時，油價往往會下降） 和高的統計意義 （詳細資訊請見附錄 2 月本文下面）。
未完
Justin Lai 編譯
03/30/2016
Justin Lai
沒有留言:
張貼留言