FedWatch聯準會目標升息幅度機率對股價影響之研究

Abstract

過去針對貨幣政策及聯邦基金利率對股價影響的研究很多，對貨幣政策決定因素的研究也不少。然而，針對FedWatch聯準會目標升息幅度機率對股價影響的研究文獻卻沒有。本研究嘗試研究FedWatch聯準會目標升息幅度機率與股價的關係及他們的決定因素。希望藉由本研究的發現提供投資人及政策制定者做更好的決策。

本研究整理FedWatch聯準會目標升息幅度機率計算工具，及目標利率預測機率，並從FedWatch網站下載1/3/2022至12/14/2022 (聯準會開會日期)該預測目標利率機率資料。相對應期間道瓊基金指數及聯邦基金利率資料則由雅虎財金資料庫取得。進一步探討FedWatch聯準會目標升息幅度機率與股價之關係。

本研究顯示股價與聯邦基金利率呈現負相關，該結果與過去相關文獻研究一致。研究中發現FedWatch聯準會目標升息幅度機率與股價之關係為顯著負相關。 亦即FedWatch聯準會目標升息幅度機率對股價有顯著負面影響，該影響隨時間落後變數遞減。此外，聯邦基金利率與FedWatch機率加權平均目標利率落差變數呈現正向遞增關係。然而近期落後變數參數估計值並不顯著，只有遠期落差變數顯著正相關，可能與FedWatch聯準會目標升息幅度機率計算與遠期利率期貨合約有關。

根據研究結論，本研究擬定三種交易策略。模擬結果顯示三種交易策略的累積日報酬均超過大盤。其中又以交易策略三累積日報酬最優且全樣本期間均超越大盤。 最後針對本研究假設，模型及交易策略就不同樣本資料期間進行安定性檢定，其結論並無改變。

Prior research on the impact of monetary policy and Fed fund rate changes on stock prices has been fully documented. Some papers discussed the determinants of monetary policy too, but the impact of FedWatch probability of FOMC meeting decision on Fed fund target rate changes has not yet been studied. This paper tries to study the relationships among the FedWatch probabilities on target fund rate, the determinants of Fed fund rate changes, and stock prices, thus providing insight to investors and implications for policymakers.

This thesis reviews FedWatch probabilities calculator tools, and the daily data of the probabilities of fed fund target rate changes, which are generated during the Federal Reserve Meetings and downloaded from the FedWatch website starting from 1/3/2022 to 12/14/2022, the meeting date. Stock indices daily data of Dow Johns Industrial Average and Fed fund rates for the same period are collected from the Yahoo Finance database. The relationships between FedWatch probabilities of fed fund target rate changes in Federal Reserve Meetings and stock prices are investigated. The results show that the relationship between the stock return and federal fund rates is negative which are consistent with those of the previous researches. We also find that the FedWatch probabilities on Fed target fund rate changes are all significantly inversely related to the stock prices. The impact of the FedWatch probabilities lag variables on the stock prices is negative at diminishing path with distribution lags. The relationships between the effective Fed fund rate and the probability-weighted average of target Fed fund rate changes are positively related. The distribution lag estimates show that FedWatch probability estimates of near-term lags are not significantly related. Only the estimates of distant lags are significantly related to effective fund rates because the lag FedWatch probabilities are calculated from fed fund futures contracts which provide information about future rates. The impact of the FedWatch probabilities lag variables on the effective fund rates is positive, and the significant level increases with the lag periods.

Based on the results, three long/short trading strategies are developed to justify my conclusions and provide investors with better insight for their investments. All the three strategies have higher CDR than the stock market index and the CDR of strategy 3 dominates that of the SMI during the whole sample period, which means strategy 3 always performs better than SMI for the whole sample period.

Robustness tests for new sample period show that the hypotheses test results still hold and the excess returns for trading strategies still exit, which do not subject to the problems of changing parameters over time.