台灣房地產景氣動向季報的發佈不但為台灣房地產市場該季景氣狀況提供一客觀參考依據，且對未來房地產市場景氣走向進行預測分析，提供房地產景氣相關資訊予社會各界參考。在對未來房地產景氣進行預測方面，是採用領先指標作為一參考標的，雖然領先指標就其特性而言為一預測未來景氣動向的參考資訊，但卻無法明確指出其所預測的期間為多長，本研究所欲探討的部分即在於使用房地產景氣綜合指標之領先指標於預測房地產景氣上可以預測幾季之後的房地產景氣狀況，並探討房地產景氣預測的準確度。另外對廠商景氣預測的部分進行一致性的檢視，再從而比較廠商景氣認知與同時、領先指標之間的關係。希望藉此能釐清房地產景氣季報內的訊息內涵。
本研究透過因果關係檢定，找出領先指標領先同時指標三季的關係，確定可以透過領先指標來預測三季之後的景氣變化情形。再利用平均絕對百分比誤差分析使用領先指標作為預測工具的準確度，經過計算求得平均絕對百分比誤差為4.57%，可以看出使用領先指標作為預測工具相當合適。在廠商本身預測景氣動向上，本研究發現廠商預期符合調適預期理論，且對未來的預期有較為樂觀的趨勢；經過本研究統計整理發現廠商對房地產景氣變動的認知與同時指標的變動之間的一致性並不理想，此有可能是因為市場上各項房地產資訊充斥，廠商並不全然以同時指標的變動來判斷房地產景氣的變動。惟經過前後期的比較，可以發現一致性的上升與不一致性的下降，表示兩者之間認知的一致性有逐漸提升的趨勢。 / The atmosphere of the real estate industry in Taiwan was upswing gradually in the past few decades, and its market information was definitely occupied an significant role and many people started to put their focus on it. In fact, the industry forecast which can be found from those seasonal report was a prediction only that is not accurate, i.e. those projections were based on the current market information only. Therefore, how to interpret and digest the number from those reports is the crucial issue for the reader to think and to consider. This study is trying to find out the time-lags between leading index and coincident index and the accuracy of the cycle forecasting system.
The Granger causality test is widely used to examine the time-lags between the leading index and the coincident index. From the result of this study, we can found that there has a 3-season time-gap embedded in, and this is an obvious difference indeed. In addition, the forecasted index of the leading indicator i.e. The MAPE is in 4.57% sharp. Thus, we dare to judge that the result of this study is absolutely accurate with 100% confidence.
| Identifer | oai:union.ndltd.org:CHENGCHI/G0091257027 |
| Creators | 詹任偉 |
| Publisher | 國立政治大學 |
| Source Sets | National Chengchi University Libraries |
| Language | 中文 |
| Detected Language | English |
| Type | text |
| Rights | Copyright © nccu library on behalf of the copyright holders |
Page generated in 0.0022 seconds