Cointegration, directed graphs and recursive structural model : an application to prices and money, U.S. data 1869-1957 /
Whether money causes movement in the general price icrofilm Inc. level or not might be better understood with a differenced vector autoregressive (VAR) model with error correction terms than with a simple regression analysis. The regression analyses of Bordo and Schwartz (1981) were not sufficient t...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1998.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=733038611&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | Whether money causes movement in the general price icrofilm Inc. level or not might be better understood with a differenced vector autoregressive (VAR) model with error correction terms than with a simple regression analysis. The regression analyses of Bordo and Schwartz (1981) were not sufficient to convince one logically that money is the cause of the movement of the general price level. The coefficients of the regression analysis only say that money and the general price level are correlated. A VAR model with general price level are correlated. A VAR model with correction terms, however, can show not only whether there are long-run stable relations between the variables, but also what the impulse responses are in the model. The meaningful impulse responses among economic time series are obtained under the reasonable order of the variables. A directed acyclic graph (DAG) model proposed by Spices, Glymour and Scheines (1993) enables us to construct the order of the variables in the model using partial correlations. The recursive structural model proposed by Swanson and Granger (1997) offers a simpler way to find the order of the variables. The Swanson and Granger method is, however, more restrictive than the DAG approach because Swanson and Granger use only first order partial correlation. Two results will be compared. In terms of Two results will be compared. In terms of five variables, which are the ratio of money to output, the general price level, relative price between wheat and coal, and the short-term and long-term interest rates, shows that the long-run relation between money and the general price level is not overwhelming when considering the extended time period of 1869-1957 compared to the time period of 1869-1914 which Bordo and Schwartz (1918) studied. |
|---|---|
| Item Description: | Vita. "Major Subject: Economics". |
| Physical Description: | xi, 140 leaves : illustrations ; 28 cm. |
| Bibliography: | Includes bibliographical references (leaves 136-139). |