Model Dinamis

Tedy Herlambang

<div class="lab-logo"></div>

# Model Dinamis

## Ekonometrika Sesi 6

### Tedy Herlambang .small[ ]

[ bangtedy.github.io](https://bangtedy.github.io)

[ @t_hlb](https://twitter.com/t_hlb)

---

# Pendahuluan

- Kebijakan ekonomi dampaknya tidak instan, ada kelambanan (lag)
- Model regresi yang memasukkan unsur lag disebut model dinamis

Dengan mempeljari modul 6, Anda diharapkan dapat:

1. Menjelaskan model autoregressive
2. Membentuk model penyesuaian parsial, mengestimasi, mengevaluasi dan mengartikan arti koefisien regresi model penyesuaian parsial
3. Membentuk parsial adjustment model (PAM) mengestimasi, mengevaluasi dan menratikan arti koefisien regresi model PAM
4. Mengkaji autokorelasi jika model dalam bentuk autoregressive
5. Membentuk, mengestimasi, mengevaluasi hasil model kausalitas Granger

---

# KB1: MODEL REGRESI DENGAN KELAMBANAN
<hr width="100%" align="left" size="0.3" color="gold"></hr>

---
# Model Kelambanan

- Pengaruh kebijakan fiskal dan moneter tidak instan --> perlu waktu 6-12 bulan
- Kebijakan ekonomi `$X_t$` mempunyai dampak ekonomi `$Y_t$`, `$Y_{t+1}$`, `$Y_{t+2}$`, dst.
- Atau jika dibalik: variabel dependen `$Y_t$` dipengaruhi oleh `$X_t$`, `$X_{t-1}$`, `$X_{t-2}$`, dst.
- Model yang memasukkan nilai current dan juga kelambanan disebut model kelambanan/*distributed-lag model*:

`$$Y_t = \alpha+\beta_0X_t+\beta_1X_{t-1}+\beta_2X_{t-2}+...+\beta_kX_{t-k}+\epsilon$$`

---

# Model Kelambanan Geometrik

- Dibentuk dari pendekatan matematis dimana setiap variabel ekonomi akan mempunyai pengaruh yang semakin mengecil

---
# Model Penyesuaian Adaptif
- Nilai harapan dimasa mendatang hanya tergantung dari informasi periode sebelumnya

---
# Partial Adjustment Model (PAM)
- Dibangun atas dasar teori penyesuaian persediaan (*stock*)
- Persediaan optimal merupakan perbedaan persediaan yang diinginkan dan persediaan periode sebelumnya.
- Suppose the adjustment of the actual value of a variable Yt to its optimal (or desired) level (denoted by Yt∗) needs to be modelled.
- Assumes that the change in actual Yt (that is, Yt − Yt−1) will
be equal to a proportion of the optimal change (Yt∗ − Yt−1), or:

`$$Y_t − Y_{t−1} = λ(Y^*_t − Y_{t-1})$$`
- λ is the adjustment coefﬁcient, which takes values from 0 to 1, and 1/λ denotes the speed of adjustment
> Perhatikan dan praktekkan contoh yang diberikan Tutor
---
# Estimasi Model Autoregresif
> Ketiga model kelambanan sebelumnya akan menghasilkan suatu model akhir yaitu model *autoregresif*

- An autoregressive equation is a model with a lagged dependent variable
---
# Deteksi Autokorelasi Model Autoregresif

---
# Model Polinomial

---
# Pemilihan Panjang Kelambanan

---
# An Example of a Dynamic Model
- let’s look at an aggregate consumption function from a macroeconomic equilibrium GDP model. 
- Many economists argue that in such a model, consumption (COt) is not just an instantaneous function of disposable income (YDt). - Instead, they believe that current consumption is also influenced by past levels of disposable income (YDt - 1, YDt - 2, etc.)

`$$CO_t = \alpha_0 + \beta_0YD_t + \beta_1YD_{t - 1} + \beta_2YD_{t -2} + ... + \beta_pYD_{t-p} + \epsilon_t$$`

- The coefficients of past levels of income decrease as the length of the lag increases. 
- The impact of lagged income on current consumption should decrease as the lag gets bigger. 
- Thus we’d expect the coefficient of YDt - 2 to be less than the coefficient of YDt - 1, and so on.

---

# KB2: MODEL KAUSALITAS
<hr width="100%" align="left" size="0.3" color="gold"></hr>

---
# Kausalitas dalam Teori Ekonomi
- Banyak variabel ekonomi menunjukkan hubungan dua arah atau kausalitas
  - nilai tukar mempengaruhi inflasi
  - inflasi mempengaruhi nilai tukar
  - Ekspor dengan GDP
  - Pertumbuhan ekonomi dengan jumlah uang beredar
  - Pengeluaran pemerintah dan penerimaan pemerintah
---
# Model Kausalitas Granger
- Regresi menunjukkan hubungan satu arah: dari variabel independen ke variabel dependen
- Kausalitas menunjukkan hubungan dua arah --> semua variabel dependen, tidak ada variabel independen
- Salah satu uji kausalitas adalah uji Granger
  - To provide evidence about the direction of causality in economic relationships. 
  - The  test is useful when we know that two variables are related but we don’t know which variable causes the other to move

---
# Granger causality

- Granger causality, or precedence, is a circumstance in which one time-series variable consistently and predictably changes before another variable.
- Granger causality is important because it allows us to analyze which variable precedes or “leads” the other.
- Despite the value of Granger causality, however, we shouldn’t let ourselves be lured into thinking that it allows us to prove economic causality in any rigorous way. 
- If one variable precedes (“Granger causes”) another, we can’t 
be sure that the first variable “causes” the other to change.
- As a result, even if we’re able to show that event A always happens before event B, we have not shown that event A “causes” event B.

> Perhatikan dan praktekkan contoh yang diberikan Tutor
---

Matur nuwun, find me at...

[ @t_hlb ](https://twitter.com/t_hlb)

[ bangtedy.github.io](https://bangtedy.github.io)