Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso...
-
Upload
international-food-policy-research-institute -
Category
Technology
-
view
658 -
download
4
description
Transcript of Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso...
![Page 1: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/1.jpg)
Predicting Time Periods of Excessive Price Volatility:The Case of Rice
Dr. Ramon Clarete Alfonso Labao
University of the Philippines, School of Economics
September 25, 2013
![Page 2: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/2.jpg)
Flow of Presentation
Overview
Methodology
Empirical Results
Conclusion and Constraints
![Page 3: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/3.jpg)
Overview:
Importance of being forewarned of extreme food price-volatility:
provides time to undertake cooperation
prevent herding and self-fulfilling crises
avoid repetition of 2008 rice crisis
prevent welfare costs for the poor
G20 report stresses importance of accurate and timely market information
![Page 4: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/4.jpg)
Overview:
Importance of being forewarned of extreme food price-volatility:
provides time to undertake cooperation
prevent herding and self-fulfilling crises
avoid repetition of 2008 rice crisis
prevent welfare costs for the poor
G20 report stresses importance of accurate and timely market information
![Page 5: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/5.jpg)
Overview:
Importance of being forewarned of extreme food price-volatility:
provides time to undertake cooperation
prevent herding and self-fulfilling crises
avoid repetition of 2008 rice crisis
prevent welfare costs for the poor
G20 report stresses importance of accurate and timely market information
![Page 6: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/6.jpg)
Overview:
Importance of being forewarned of extreme food price-volatility:
provides time to undertake cooperation
prevent herding and self-fulfilling crises
avoid repetition of 2008 rice crisis
prevent welfare costs for the poor
G20 report stresses importance of accurate and timely market information
![Page 7: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/7.jpg)
Our Methodology:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 8: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/8.jpg)
Our Methodology:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 9: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/9.jpg)
Our Methodology:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 10: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/10.jpg)
Our Methodology:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 11: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/11.jpg)
First Two Parts:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 12: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/12.jpg)
Actual Prices:
![Page 13: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/13.jpg)
Snapshot of actual prices:
![Page 14: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/14.jpg)
Converted into Price Returns:
![Page 15: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/15.jpg)
Converted into Price Returns:
where price returns refer to day-to-day price movements, or:
price return = ln(price of today
price of yesterday)
![Page 16: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/16.jpg)
Converted into Price Returns:
where price returns refer to day-to-day price movements, or:
price return = ln(price of today
price of yesterday)
![Page 17: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/17.jpg)
Converted into Price Returns:
where price returns refer to day-to-day price movements, or:
price return = ln(price of today
price of yesterday)
![Page 18: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/18.jpg)
Create Thresholds via MTY’s two-step methodology:
![Page 19: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/19.jpg)
Create Thresholds via MTY’s two-step methodology:
![Page 20: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/20.jpg)
Create Thresholds via MTY’s two-step methodology:
![Page 21: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/21.jpg)
Methodology:
MTY’s Two-step method:
Construct a nonparametric trend via spline-backfitted-kernel:
rt = m0 +d∑
a=1
ma(Xta) + (h0 +d∑
a=1
ha(Xta)1/2)εt
Estimate the high-order 95 percent quantile via Generalized Pareto Distribution (GPD):
q̂t(α) = ε̂k+1 +β̂tγ̂t
(((1 − α)
(k/N))−γ̂t − 1)
Estimated 95 percent Conditional Quantiles:
q̂t(α/rt−1, rt−2) = m̃(rt−1, rt−2) + [(h̃(rt−1, rt−2))1/2q̂t(α)]
![Page 22: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/22.jpg)
Methodology:
MTY’s Two-step method:
Construct a nonparametric trend via spline-backfitted-kernel:
rt = m0 +d∑
a=1
ma(Xta) + (h0 +d∑
a=1
ha(Xta)1/2)εt
Estimate the high-order 95 percent quantile via Generalized Pareto Distribution (GPD):
q̂t(α) = ε̂k+1 +β̂tγ̂t
(((1 − α)
(k/N))−γ̂t − 1)
Estimated 95 percent Conditional Quantiles:
q̂t(α/rt−1, rt−2) = m̃(rt−1, rt−2) + [(h̃(rt−1, rt−2))1/2q̂t(α)]
![Page 23: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/23.jpg)
Methodology:
MTY’s Two-step method:
Construct a nonparametric trend via spline-backfitted-kernel:
rt = m0 +d∑
a=1
ma(Xta) + (h0 +d∑
a=1
ha(Xta)1/2)εt
Estimate the high-order 95 percent quantile via Generalized Pareto Distribution (GPD):
q̂t(α) = ε̂k+1 +β̂tγ̂t
(((1 − α)
(k/N))−γ̂t − 1)
Estimated 95 percent Conditional Quantiles:
q̂t(α/rt−1, rt−2) = m̃(rt−1, rt−2) + [(h̃(rt−1, rt−2))1/2q̂t(α)]
![Page 24: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/24.jpg)
How are Time Periods of Excessive Price Volatility (EPV) defined?
Martins-Filho, Maximo Torero and Feng Yao’s definition of time periods of EPV:
EPV Definition:
Time-Periods whereby the preceding 60 days experienced a significantly high amountof extreme positive price returns over the 95 percent conditional quantile.
... at least 7 instances of extreme positive price returns within a span of 60 days
![Page 25: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/25.jpg)
How are Time Periods of Excessive Price Volatility (EPV) defined?
Martins-Filho, Maximo Torero and Feng Yao’s definition of time periods of EPV:
EPV Definition:
Time-Periods whereby the preceding 60 days experienced a significantly high amountof extreme positive price returns over the 95 percent conditional quantile.
... at least 7 instances of extreme positive price returns within a span of 60 days
![Page 26: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/26.jpg)
How are Time Periods of Excessive Price Volatility (EPV) defined?
Martins-Filho, Maximo Torero and Feng Yao’s definition of time periods of EPV:
EPV Definition:
Time-Periods whereby the preceding 60 days experienced a significantly high amountof extreme positive price returns over the 95 percent conditional quantile.
... at least 7 instances of extreme positive price returns within a span of 60 days
![Page 27: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/27.jpg)
1st (Jul 1993 - Feb 1994) and 2nd (Jul 1994 - Oct 1994) EPV cluster:
3rd (Mar 1999 - Sep 1999) and 4th (Jun 2001 - Aug 2001) EPV cluster:
![Page 28: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/28.jpg)
1st (Jul 1993 - Feb 1994) and 2nd (Jul 1994 - Oct 1994) EPV cluster:
3rd (Mar 1999 - Sep 1999) and 4th (Jun 2001 - Aug 2001) EPV cluster:
![Page 29: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/29.jpg)
5th (Jul 2002 - Aug 2002) and 6th (May 2003 - Oct 2003) EPV cluster:
7th (Mar 2008 - Nov 2008) and 8th (Apr 2009 - May 2009) EPV cluster:
![Page 30: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/30.jpg)
5th (Jul 2002 - Aug 2002) and 6th (May 2003 - Oct 2003) EPV cluster:
7th (Mar 2008 - Nov 2008) and 8th (Apr 2009 - May 2009) EPV cluster:
![Page 31: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/31.jpg)
9th (Jul 2011) EPV cluster:
![Page 32: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/32.jpg)
In Summary:
501 days of excessive price-volatility (EPV)
Nine (9) clusters of EPV
... Of the above clusters of EPV, only four (4) are severe
![Page 33: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/33.jpg)
In Summary:
501 days of excessive price-volatility (EPV)
Nine (9) clusters of EPV
... Of the above clusters of EPV, only four (4) are severe
![Page 34: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/34.jpg)
In Summary:
501 days of excessive price-volatility (EPV)
Nine (9) clusters of EPV
... Of the above clusters of EPV, only four (4) are severe
![Page 35: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/35.jpg)
In Summary:
501 days of excessive price-volatility (EPV)
Nine (9) clusters of EPV
... Of the above clusters of EPV, only four (4) are severe
![Page 36: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/36.jpg)
Third Part:
- Use daily rice future prices
(5407 days from Sep 1991 to Mar 2013)
- Identify time-periods of excessive price volatility
(via Martins-Filho, Maximo Torero, and Feng Yao’s (IFPRI, 2010) methodology:Two-step spline-backfitted-kernel estimation / GPD distribution)
- Develop Early Warning Signals
(via Particle Swarm Optimization (PSO))
![Page 37: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/37.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 38: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/38.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 39: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/39.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 40: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/40.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 41: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/41.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 42: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/42.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 43: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/43.jpg)
Next Task: Looking for a Good Early Warning Signal
We set the mirror-image’s lag at 60 days prior to an EPV cluster...
![Page 44: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/44.jpg)
Next Task: Looking for a Good Early Warning Signal
We set the mirror-image’s lag at 60 days prior to an EPV cluster...
![Page 45: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/45.jpg)
Next Task: Looking for a Good Early Warning Signal
![Page 46: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/46.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 47: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/47.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 48: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/48.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 49: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/49.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 50: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/50.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 51: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/51.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 52: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/52.jpg)
Next Task: Looking for a Good Early Warning Signal
Parameters of the Early Warning Signals:
Window of Observation
Lower-Order Quantile Level
Frequency of Breach
Scope
Together, these parameters generate a time-based variable, namely...
![Page 53: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/53.jpg)
Next Task: Looking for a Good Early Warning Signal
... the scope’s time-coverage
![Page 54: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/54.jpg)
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
Basically...
We create a new set of trends (using spbk) and solve for parameters that meet aboverequirements
![Page 55: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/55.jpg)
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
Basically...
We create a new set of trends (using spbk) and solve for parameters that meet aboverequirements
![Page 56: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/56.jpg)
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
Basically...
We create a new set of trends (using spbk) and solve for parameters that meet aboverequirements
![Page 57: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/57.jpg)
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
Basically...
We create a new set of trends (using spbk) and solve for parameters that meet aboverequirements
![Page 58: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/58.jpg)
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
Basically...
We create a new set of trends (using spbk) and solve for parameters that meet aboverequirements
![Page 59: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/59.jpg)
We expressed the criteria (accurate, comprehensive, good lead time), into a singleadditive objective function that can be minimized:
Objective Function for Minimization:
f (X ) = [(count1
laghvp) ∗ 1.35] + [
count2
hvp] + [(accuracy) ∗ 1.35]+
[(totalews
totaldays) ∗ 1.25] + [
totalscope
totaldays]
Where:
count1: total no. of lagged mirror-images of time periods of EPV not covered bythe scopes of the signals
laghvp: total no. of lagged mirror-images of time periods of EPV
count2: total no. of actual time periods of EPV not covered by the scopes’time-coverage of the early warning signals
hvp: total no. of actual time periods of EPV
accuracy: predictive accuracy of the early warning signal with following ratio:[1-((no.of high.vol.periods captured by scope’s time-coverage) / (hvp)].
total ews: total no. of early warning signals
total scope: total no. of days covered by the scopes’ time-coverage of the earlywarning signals
total days: total no. of days within the sample timeframe (1991 to 2013): 5407future days
![Page 60: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/60.jpg)
We expressed the criteria (accurate, comprehensive, good lead time), into a singleadditive objective function that can be minimized:
Objective Function for Minimization:
f (X ) = [(count1
laghvp) ∗ 1.35] + [
count2
hvp] + [(accuracy) ∗ 1.35]+
[(totalews
totaldays) ∗ 1.25] + [
totalscope
totaldays]
Where:
count1: total no. of lagged mirror-images of time periods of EPV not covered bythe scopes of the signals
laghvp: total no. of lagged mirror-images of time periods of EPV
count2: total no. of actual time periods of EPV not covered by the scopes’time-coverage of the early warning signals
hvp: total no. of actual time periods of EPV
accuracy: predictive accuracy of the early warning signal with following ratio:[1-((no.of high.vol.periods captured by scope’s time-coverage) / (hvp)].
total ews: total no. of early warning signals
total scope: total no. of days covered by the scopes’ time-coverage of the earlywarning signals
total days: total no. of days within the sample timeframe (1991 to 2013): 5407future days
![Page 61: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/61.jpg)
We expressed the criteria (accurate, comprehensive, good lead time), into a singleadditive objective function that can be minimized:
Objective Function for Minimization:
f (X ) = [(count1
laghvp) ∗ 1.35] + [
count2
hvp] + [(accuracy) ∗ 1.35]+
[(totalews
totaldays) ∗ 1.25] + [
totalscope
totaldays]
First two (2) terms - optimize good lead-time and comprehensiveness
Last three (3) terms - minimize false alarms and improve accuracy.
![Page 62: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/62.jpg)
We expressed the criteria (accurate, comprehensive, good lead time), into a singleadditive objective function that can be minimized:
Objective Function for Minimization:
f (X ) = [(count1
laghvp) ∗ 1.35] + [
count2
hvp] + [(accuracy) ∗ 1.35]+
[(totalews
totaldays) ∗ 1.25] + [
totalscope
totaldays]
First two (2) terms - optimize good lead-time and comprehensiveness
Last three (3) terms - minimize false alarms and improve accuracy.
![Page 63: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/63.jpg)
We expressed the criteria (accurate, comprehensive, good lead time), into a singleadditive objective function that can be minimized:
Objective Function for Minimization:
f (X ) = [(count1
laghvp) ∗ 1.35] + [
count2
hvp] + [(accuracy) ∗ 1.35]+
[(totalews
totaldays) ∗ 1.25] + [
totalscope
totaldays]
First two (2) terms - optimize good lead-time and comprehensiveness
Last three (3) terms - minimize false alarms and improve accuracy.
![Page 64: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/64.jpg)
We use R’s PSO package to minimize the objective function...
PSO: Particle Swarm Optimization
PSO is a meta-heuristic designed for functions that are hard to optimize usingtraditional optimization procedures (due to several peaks, non-linearities, long-forms,etc..)
PSO Parameter Search Range:
Window of Observation: from 1 to 40 future days
Quantile Level: from 50 percent to 99 percent quantile
Frequency of Breach: from 1 to 5
Scope of Early Warning Signal: from 60 to 250 future days
![Page 65: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/65.jpg)
We use R’s PSO package to minimize the objective function...
PSO: Particle Swarm Optimization
PSO is a meta-heuristic designed for functions that are hard to optimize usingtraditional optimization procedures (due to several peaks, non-linearities, long-forms,etc..)
PSO Parameter Search Range:
Window of Observation: from 1 to 40 future days
Quantile Level: from 50 percent to 99 percent quantile
Frequency of Breach: from 1 to 5
Scope of Early Warning Signal: from 60 to 250 future days
![Page 66: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/66.jpg)
We use R’s PSO package to minimize the objective function...
PSO: Particle Swarm Optimization
PSO is a meta-heuristic designed for functions that are hard to optimize usingtraditional optimization procedures (due to several peaks, non-linearities, long-forms,etc..)
PSO Parameter Search Range:
Window of Observation: from 1 to 40 future days
Quantile Level: from 50 percent to 99 percent quantile
Frequency of Breach: from 1 to 5
Scope of Early Warning Signal: from 60 to 250 future days
![Page 67: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/67.jpg)
We use R’s PSO package to minimize the objective function...
PSO: Particle Swarm Optimization
PSO is a meta-heuristic designed for functions that are hard to optimize usingtraditional optimization procedures (due to several peaks, non-linearities, long-forms,etc..)
PSO Parameter Search Range:
Window of Observation: from 1 to 40 future days
Quantile Level: from 50 percent to 99 percent quantile
Frequency of Breach: from 1 to 5
Scope of Early Warning Signal: from 60 to 250 future days
![Page 68: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/68.jpg)
After 1000 iterations, we obtain the following optimal solution:
Optimal Solution:
Window of Observation: 11.61 future days
Quantile Level: 91.31 percent
Frequency of Breach: 4.18
Scope of Early Warning Signal: 107.06 future days
Given these parameters, an early warning signal will be activated...
... once the 91.31 percent conditional quantile is breached five (5) times within11.61 future days..
![Page 69: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/69.jpg)
After 1000 iterations, we obtain the following optimal solution:
Optimal Solution:
Window of Observation: 11.61 future days
Quantile Level: 91.31 percent
Frequency of Breach: 4.18
Scope of Early Warning Signal: 107.06 future days
Given these parameters, an early warning signal will be activated...
... once the 91.31 percent conditional quantile is breached five (5) times within11.61 future days..
![Page 70: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/70.jpg)
After 1000 iterations, we obtain the following optimal solution:
Optimal Solution:
Window of Observation: 11.61 future days
Quantile Level: 91.31 percent
Frequency of Breach: 4.18
Scope of Early Warning Signal: 107.06 future days
Given these parameters, an early warning signal will be activated...
... once the 91.31 percent conditional quantile is breached five (5) times within11.61 future days..
![Page 71: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/71.jpg)
After 1000 iterations, we obtain the following optimal solution:
Optimal Solution:
Window of Observation: 11.61 future days
Quantile Level: 91.31 percent
Frequency of Breach: 4.18
Scope of Early Warning Signal: 107.06 future days
Given these parameters, an early warning signal will be activated...
... once the 91.31 percent conditional quantile is breached five (5) times within11.61 future days..
![Page 72: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/72.jpg)
Just a review...
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
![Page 73: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/73.jpg)
Just a review...
What makes a good early warning signal:
accurate: scope’s time-coverage covers an EPV cluster, minimal false alarms
good lead time: there’s reasonable lead time before EPV cluster
comprehensive: signals pre-empt almost all EPV clusters
![Page 74: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/74.jpg)
We obtain these Performance Statistics:
accuracy of ews: 94.44 percent predictive power
comprehensiveness: 99.40 percent of time periods of EPV are covered
lead-time before EPV cluster: average of 43 days from the earliest signal to start ofhigh-volatility time-period
![Page 75: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/75.jpg)
We obtain these Performance Statistics:
accuracy of ews: 94.44 percent predictive power
comprehensiveness: 99.40 percent of time periods of EPV are covered
lead-time before EPV cluster: average of 43 days from the earliest signal to start ofhigh-volatility time-period
![Page 76: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/76.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Total of 125 early warning signals activated from Sep 1991 to Mar 2013.
This is only 2.3 percent of the time.
![Page 77: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/77.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Total of 125 early warning signals activated from Sep 1991 to Mar 2013.
This is only 2.3 percent of the time.
![Page 78: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/78.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Total of 125 early warning signals activated from Sep 1991 to Mar 2013.
This is only 2.3 percent of the time.
![Page 79: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/79.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Average of 43 days between earliest ews and start of an EPV cluster
But among the four (4) severe EPV clusters, there is an average lead-time of 61days
The unflagged 2011 EPV cluster is not severe..
All EPV are sufficiently covered from beginning till end by the scopes of theews..
![Page 80: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/80.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Average of 43 days between earliest ews and start of an EPV cluster
But among the four (4) severe EPV clusters, there is an average lead-time of 61days
The unflagged 2011 EPV cluster is not severe..
All EPV are sufficiently covered from beginning till end by the scopes of theews..
![Page 81: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/81.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Average of 43 days between earliest ews and start of an EPV cluster
But among the four (4) severe EPV clusters, there is an average lead-time of 61days
The unflagged 2011 EPV cluster is not severe..
All EPV are sufficiently covered from beginning till end by the scopes of theews..
![Page 82: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/82.jpg)
Below is a summary of the different EPV clusters and the lead-times of the ews:
Average of 43 days between earliest ews and start of an EPV cluster
But among the four (4) severe EPV clusters, there is an average lead-time of 61days
The unflagged 2011 EPV cluster is not severe..
All EPV are sufficiently covered from beginning till end by the scopes of theews..
![Page 83: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/83.jpg)
for illustration...
![Page 84: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/84.jpg)
1st (Jul 1993 - Feb 1994) and 2nd (Jul 1994 - Sep 1994) EPV Clusters:
3rd EPV Cluster (Mar 1999 - Sep 1999)and 4th (Jun 2001 - Aug 2001) EPV Clusters:
![Page 85: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/85.jpg)
1st (Jul 1993 - Feb 1994) and 2nd (Jul 1994 - Sep 1994) EPV Clusters:
3rd EPV Cluster (Mar 1999 - Sep 1999)and 4th (Jun 2001 - Aug 2001) EPV Clusters:
![Page 86: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/86.jpg)
5th (Jul 2002 - Aug 2002) and 6th (May 2003 - Oct 2003) EPV Clusters:
7th (Mar 2008 - Nov 2008) and 8th (Apr 2009 - May 2009) EPV Clusters:
![Page 87: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/87.jpg)
5th (Jul 2002 - Aug 2002) and 6th (May 2003 - Oct 2003) EPV Clusters:
7th (Mar 2008 - Nov 2008) and 8th (Apr 2009 - May 2009) EPV Clusters:
![Page 88: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/88.jpg)
5th (Jul 2002 - Aug 2002) and 6th (May 2003 - Oct 2003) EPV Clusters:
7th (Mar 2008 - Nov 2008) and 8th (Apr 2009 - May 2009) EPV Clusters:
![Page 89: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/89.jpg)
9th (Jul 2011) EPV Cluster:
![Page 90: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/90.jpg)
Constraints:
usage of daily future prices, not daily spot prices
sensitivity analysis
application to other commodities
![Page 91: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/91.jpg)
Constraints:
usage of daily future prices, not daily spot prices
sensitivity analysis
application to other commodities
![Page 92: Predicting Time Periiods of Excessive Price Volatility: The Case of Rice- Ramon Clarete and Alfonso Labao](https://reader031.fdocuments.net/reader031/viewer/2022020206/546c18dab4af9f892c8b4f6d/html5/thumbnails/92.jpg)
Thank You.