Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
jxin2618 authored Nov 17, 2021
1 parent 026a5ff commit b50798e
Showing 1 changed file with 7 additions and 5 deletions.
12 changes: 7 additions & 5 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,6 @@ Cost of carry model; Vector error correction model; Engle-Grangle test; Johansen
- CU.xlsx: contains the data of SHFE copper
- in four levels: cash price, future price, inventory, interest rate of government bonds.
- in the historical period of 2009/12/31 - 2020/07/03
-
## Project Research

## Project Introduction
Expand Down Expand Up @@ -75,13 +74,16 @@ $$ DP_t = a_0 + a_1 DF_{t|t-1} + e_t \tag{1} $$
$$ DP_t = b_0 + b_1 (F_{t|t-1} - P_{t-1}) + e_t \tag{2} $$
- Brenner and Kroner Model(1995):
$$ DP_t = g_0 + g_1 DF_{t|t-1} + g_2 Dr_{t|t-1} + g_3 DI_{t-1} + g_4 D\sigma_{t-1}+$$
$$ g_5 D\rho(1)_{t-1} + g_6ECT_{t-2} + e_t \tag{3} $$
$$ g_5 D\rho (1)_{t-1} + g_6ECT_{t-2} + e_t \tag{3} $$

The error correction term, $ECT_{t-2}$ is the residual, $e_{t-1}$ from the regression:
$$ P_{t-1} = h_0 + h_1 F_{t-1|t-2} + h_2 r_{t-1|t-2} + \newline h_3I_{t-2} +h_4 \sigma_{t-2} + h_5\rho(1)_{t-2} + e_{t-1} $$
$$ P_{t-1} = h_0 + h_1 F_{t-1|t-2} + h_2 r_{t-1|t-2} + h_3I_{t-2} +h_4 \sigma_{t-2} + h_5\rho (1)_{t-2} + e_{t-1} $$

- Vector Error Correction Model:
$$ DX_t = M_1 DX_{t-1} + S^{'}ECT_{t-2}^{*} + L + E_t \tag{4} $$
$DX_t$ is the change in vector $X$, $M_1$ is the matrix of parameters, $S$ is the vector of speed of adjustment parameters, $L$ is a vector of constant terms and $E_t$ is the vector of residuals. $ECT^*_{t-2}$ are the residuals, $e_t$ from the following regression.
$$ P_{t-2} = j_0 + j_1 F_{t-1|t-2} + j_2 r_{t-1|t-2} + j_3I_{t-2} + j_4 \sigma_{t-2} + j_5\rho(1)_{t-2} + e_{t-1}$$

$DX_t$ is the change in vector $X$, $M_1$ is the matrix of parameters, $S$ is the vector of speed of adjustment parameters, $L$ is a vector of constant terms and $E_t$ is the vector of residuals.$ECT^{*}_{t-2}$ are the residuals, $e_t$ from the following regression.

$$ P_{t-2} = j_0 + j_1 F_{t-1|t-2} + j_2 r_{t-1|t-2} + j_3I_{t-2} + j_4 \sigma_{t-2} + j_5\rho(1)_{t-2} + e_{t-1} $$

with $i_i$ and $j_i$ as estimated parameters.

0 comments on commit b50798e

Please sign in to comment.