Presentation continuous series

May 25, 20130 likes747 views

This document discusses different methods for constructing continuous futures price series from discrete contracts for analysis and back-testing of trading strategies. It describes constant month series, constant distance/rolling nearby series, time to maturity weighted series, and volume/OI weighted series. The document provides an example of how a time to maturity weighted series is calculated and shows it produces a smoother price representation compared to a nearby series. It also compares the return distributions and strengths/weaknesses of nearby, weighted, and generic Bloomberg series.

Continuous Future Series
Construction
Quant Team
Ruchi Agri-Trading
Singapore
May 03, 2013

Why Continuous Future Series?
• We require a continuous historical price series for
analysis or back-testing of trading strategies.
• One difficulty with the futures historical data is
associated with splicing contracts together at
contract boundaries.
• The results of any analysis will depend on the
method we choose to construct the data series.

Different types of Continuous Series
• Constant Month Series (S Nov1, etc.)
• Constant Distance / Rolling Nearby / N-th
Nearest Contract Series (S 1, BO2, etc.)
• Time to maturity weighted Series (Risk Metrics
linear interpolation approach)
• Volume/OI weighted Series

Next Nearby/Rolling Nearby
13.6
13.8
14
14.2
14.4
14.6
14.8
15
15.2
Next Nearby Series
S.X2
S.F3
Next.Nearby

Time to Maturity Weighted Series
(RiskMetrics Method)

An Example: Time to Maturity
Weighted Series
Dates
Contract Expiration
Number of Days to
Expiration
Weights (%) Actual Contract Prices
CMF
Price
NB1 1m CMF NB2 NB1
1m
CMF
NB2 ࣓ࡺ࡮૚ ࣓ࡺ࡮૛ ࡼࡶ࢛࢒ ࡼ࡭࢛ࢍ ࡼࡿࢋ࢖ ࡼࢉ࢓ࢌ
2011-
07-11
07-14 08-10 08-12 3 30 32 0.10 0.90 13.5250 13.4650 13.4450 13.4710
2011-
07-12
07-14 08-11 08-12 2 30 31 0.07 0.93 13.6650 13.5550 13.5500 13.5623
2011-
07-13
07-14 08-12 08-12 1 30 30 0.03 0.97 13.8700 13.7475 13.7475 13.7516
2011-
07-14
07-14 08-15 08-12 0 30 29 0.00 1.00 13.9050 13.8200 13.7975 13.8200
2011-
07-15
08-12 08-15 09-14 28 30 61 0.93 0.07 - 13.8575 13.8300 13.8557
2011-
07-18
08-12 08-17 09-14 25 30 58 0.83 0.17 - 13.8550 13.8225 13.8496

Time to Maturity Weighted Series
13.5
14
14.5
15
15.5
16
Constant Distance Weighted Series
S.X2
S.F3
Weighted

Comparison - Different Methods
0
20000
40000
60000
80000
100000
120000
140000
7
8
9
10
11
12
13
14
15
16
1/8/2008 1/9/2008 1/10/2008 1/11/2008 1/12/2008
Comparion b/w Weighted/Unweighted/Generic Bloomberg Series
Nearby(Actual)
Weighted
Generic S1
VOLUME

Comparison - Different Methods
-3
-2.5
-2
-1.5
-1
-0.5
0
0.5
1
2/1/2008 2/3/2008 2/5/2008 2/7/2008 2/9/2008 2/11/2008
Difference b/w Weighted Series & Generic BBG Series
BBG Generic - Weighted

Comparison - Return Distributions
Next Nearby Weighted
Std Dev 0.018 0.016
Skewness -0.905 -0.331
Kurtosis 20.612 4.860

Comparison - Different Methods
Method Strengths Weaknesses
Nth Nearest contract No data adjustment. Represents
actual values for today’s market.
Gaps in historical data.
Weighted Series Smoothens in data series. Good
for statistical research.
Non-tradable price
representations.
Constant Month
Series
Offers a long period of time in
historical data on an annualized
basis without adjusting the data.
Small Open Interest/volume
periods

This document summarizes a study modeling dynamic behavior of daily soybean prices using a GARCH-Jump model. The model was estimated using 20 years of CBOT soybean futures data and found evidence of jumps in both returns and volatility. Parameter estimates were provided and results showed the jump component accounted for around 16% of total variance. Applications of the model for option pricing, simulation, and risk analysis were discussed.

Ras mrcAmit Sinha

��

Market risk capital (MRC) is derived from value-at-risk (VaR) and a tail-loss penalty based on backtesting results. The Basel II approach to calculating the tail-loss penalty (K-factor) does not adequately account for the magnitude of losses beyond VaR. An alternative approach by Blanco-Ihle defines tail losses as a percentage breach from VaR conditional on a breach. This allows calculating the K-factor based on either the maximum tail loss (KBI.max) or the average tail losses (KBI.mean). Backtesting results for RAS portfolios in early 2012 show the KBI.mean approach provides a more realistic assessment of tail risk compared to the Basel

IRJET- Segregation of Machines According to the Noise Emitted by Differen...IRJET Journal

��

The document discusses clustering machines according to the noise they emit using hierarchical clustering. Noise data was collected from 10 machines using parameters like LAeq, LAV, LAE, and TWA. Hierarchical clustering with Johnson's algorithm was applied using both single-link and complete-link methods. For single-link, the machines clustered into groups of (1,9,7,8,4,6), (2,5), 3, and 10. For complete-link, the clustering resulted in different groupings.

Kanban for Fixed Price ProjectsSudipta Lahiri

��

This document discusses adapting kanban methods to fixed-size projects. It proposes using earned value management (EVM) techniques like tracking planned value, earned value, and actual costs to monitor project progress. Cumulative flow diagrams (CFDs) are recommended to visualize work flow and ensure the team is delivering at the desired throughput. When scope changes occur, multiple options should be evaluated using CFDs to determine their impact on capacity, timeline, and planned value before agreeing on an approach with stakeholders.

IRJET- Analysis of Crucial Oil Gas and Liquid Sensor Statistics and Productio...IRJET Journal

��

This document discusses using an autoregressive model (ARIMA) to forecast sensor statistics and production values in the oil, gas, and liquid industries based on time series data from industrial internet of things (IIOT) devices. The authors apply an ARIMA model to temperature sensor data from an oil and gas well to predict future temperature readings. They preprocess the time series data to make it stationary, identify the AR and MA model orders, fit the ARIMA model, and evaluate the accuracy of the predictions. The ARIMA model provides a way to forecast sensor values and detect potential equipment issues before failures occur based on IIOT data in oil and gas production.

Reliability Analysis of Concrete Filled Steel Tube Columns by FOSM and Subset...IRJET Journal

��

This document presents a reliability analysis of concrete filled steel tube columns using the first order second moment (FOSM) method and subset simulation technique. Experimental results are obtained for concrete filled steel tube columns with varying diameters, thicknesses, lengths, concrete grades and steel yields. A reliability analysis is conducted on a sample column and the reliability index is calculated using subset simulation. Reliability increases with diameter and concrete grade. The maximum reliability of 93.57% is obtained using the Taguchi method for specific diameter, thickness and concrete grade combinations.

Caiso MRTU implementation workshop nov 2008Kevin Forbes

��

Faculty of Engineering & Technology , Gurukula Kangri University , Haridwarashwini kumar

��

Forecasting_Quantitative Forecasting.pptRituparnaDas584083

��

Project Portfolio Francois BourcierFrancois Bourcier

��

The document outlines a project to migrate legacy systems' data and positions to a new Murex platform. It describes various deliverables and phases of the project including requirements gathering, design, build, migration, testing and go-live stages. Key activities included scoping interfaces, designing data migrations, building integrations, migrating and validating data, performing system and user acceptance testing, and conducting phased go-lives for different asset types.

Forecasting Techniquesguest865c0e0c

��

Forecasting TechniquesAnand Subramaniam

��

293749024-Reliability-Growth.pptSreekanthMylavarapu1

��

This document discusses reliability growth modeling and analysis. It defines reliability development testing and describes the test, analyze, and fix (TAAF) process used to achieve reliability growth. The document focuses on the Duane reliability growth model, explaining how it models mean time between failures (MTBF) and failure rate as a function of cumulative test time using parameters like growth rate. Graphs demonstrate how Duane models can fit reliability data and be used for reliability growth assessment and forecasting.

Forecasting.pptDrMahmoudElmarhoumy

��

The document discusses quantitative forecasting methods, including exponential smoothing. Exponential smoothing assigns weights to past observations, with more recent observations receiving greater weight, to generate forecasts. Examples are provided to demonstrate how to calculate forecasts for upcoming time periods using exponential smoothing with different smoothing constants (a). Managers can use exponential smoothing on historical demand data to forecast future demand and support inventory and production planning.

Product Design Forecasting Techniquesision.pptavidc1000

��

The document discusses quantitative forecasting methods, including naive, moving average, and exponential smoothing approaches. It provides examples of how to calculate forecasts for upcoming time periods using 3-period simple moving averages, weighted moving averages with decreasing weights, and exponential smoothing with different smoothing constants. The examples demonstrate how to iteratively calculate the forecasts for each upcoming period in a time series.

Advanced Econometrics L7-8.pptxakashayosha

��

This document provides an overview of a panel data analysis course, including: - The schedule, which covers topics like panel data models, endogeneity, time series models, and difference-in-differences. - An explanation of panel data, which has multiple cross-sections observed over time, and how it can be balanced or unbalanced. - Examples of how panel data is structured and why it is useful for studying dynamic changes and effects not seen in cross-section or time series data alone. - Methods for estimating panel data models, ranging from simple pooled OLS to fixed effects and random effects models.

Wikifolio AnalyseChristopher Helm

��

IRJET- Shock Absorber Testing using FFT Analyzer with DewesoftIRJET Journal

��

1) The document describes testing of shock absorbers using an FFT analyzer to analyze dynamic characteristics under varying conditions like road texture and spring stiffness. 2) A test rig was designed that could subject one end of the suspension to vibrations simulating road bumps while the other end experienced loading simulating vehicle and passenger weight. 3) Shock absorbers from two motorcycles (Hero Splendor and Honda Shine) were tested on the rig under different spring stiffnesses, oil types, loads, and speeds to determine optimum motion transmissibility.

IRJET- Reliability Analysis of CFST Columns by using Form and Latin Hyperc...IRJET Journal

��

This document summarizes a study on the reliability analysis of concrete filled steel tube (CFST) columns using the First Order Reliability Method (FORM) and Latin Hypercube Method (LHM). Finite element analysis was performed using ANSYS to determine ultimate loads for CFST columns with varying diameters, thicknesses, and length-to-diameter ratios. Reliability indices (β) and probabilities of failure were then calculated using the LHM in 2R rel software. The results show that increasing the steel tube thickness and decreasing the diameter-to-thickness ratio increases the ultimate load capacity and reliability of the CFST columns.

Time seriesVNRacademy

��

introduction to forecasting in operations management Chapter 3.pptsyedalirazakazmi16

��

The document discusses various forecasting techniques used in operations management. It defines forecasting as a statement about the future that helps managers plan systems and their use. Common forecasting techniques discussed include judgmental forecasts based on qualitative inputs, time series forecasts based on quantitative historical data, and associative models using explanatory variables. Specific forecasting methods covered include naive forecasts, simple and weighted moving averages, exponential smoothing, and linear regression for trend analysis. The document provides examples and explanations of how to implement these various forecasting techniques.

Webcast Presentation CPFL Energia_3Q14CPFL RI

��

- CPFL Energia reported financial results for 3Q14 with net revenue increasing 9.3% compared to 3Q13, though EBITDA decreased 19.3% due to non-recurring expenses. - Key drivers of increased net revenue included commercialization and services as well as conventional generation and CPFL Renováveis. Energy costs also increased. - Non-recurring expenses included exposure to energy purchases and reversal of provisions, totaling R$123 million in 3Q14 compared to R$28 million in 3Q13. - Adjusted for non-recurring items and proportionate consolidation of generation projects, EBITDA increased 10.4% in 3Q14 compared to the prior year period.

30120140504002IAEME Publication

��

This document analyzes factors that influence the optimization of cylindrical gear transmitters. It examines safety factors for pitting and bending based on center distance and gear ratio. Safety factor for pitting increases with center distance and decreases with gear ratio. Safety factor for bending also increases with center distance and decreases with gear ratio. The analysis also looks at safety factor for pitting based on safety factor for bending, gear ratio, and material. Safety factor for pitting increases with safety factor for bending and gear ratio. The material also influences the safety factors. Charts are provided showing the relationships between these factors. The conclusion is that center distance, gear ratio, and material all impact the safety factors and optimization of gear transmitters.

THE ANALYSIS OF SOME FACTORS FOR THE OPTIMIZATION OF CYLINDRICAL GEAR TRANSMI...IAEME Publication

��

On the phase of dimensioning of the gear transmitters have influence center distance, gear ratio, material, thermal processing etc. The purpose of this paper is analysis of factors that have influence on the optimization of cylindrical gear transmitters. It is analyzed safety factor for pitting and safety factor for bending for gear transmitters on the dependence of center distance and gear ratio, than it is analyzed safety factor for pitting on dependence on safety factor for bending, gear ratio and material.

Engineering Science WorksheetsTENSILE TESTER KIT ES6TE.docxSALU18

��

Engineering Science Worksheets TENSILE TESTER KIT ES6 TENSILE TESTS Name Date Course/Year/Group © TecQuipment Ltd 2012 Do not reproduce or transmit this document in any form or by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system without the express permission of TecQuipment Limited. TecQuipment allows you to print and photocopy this document only for use with the Engineering Science Kits. DB/1012 Workbook Page 1 of 12 ES6 Worksheet Blank Page for Notes Blank Page for Notes Workbook Page 2 of 12 ES6 Worksheet Tensile Tests Tensile Tests Aims To investigate the tensile properties of different materials. Experiment Setup As shown in Figure 1, fix the Tensile Tester to the Work Panel: Figure 1 Experiment Setup Workbook Page 3 of 12 12 ES6 Worksheet Tensile Tests Experiment Procedure 1. Read the Guidance Notes (supplied). 2. Choose your specimen and use the Dial Caliper to measure the thickness and width at the gauge length and find its cross-sectional area. For reference, measure and record the original length. 3. Fit the specimen to the Tensile Tester and set the Dial Indicator to zero. 4. Note the reading on the Vertical Scale. 5. As shown in the blank Results Table, slowly turn the Load Nut clockwise in small 0.2 mm steps up to 5 mm, then larger 1 mm and 10 mm steps until the specimen breaks. Keep the applied rate consistent (for example: take five seconds to turn the Load Nut through each 0.2 mm and then take five seconds to record the reading). At each step, record the Dial Indicator value. 6. To check the elongation, remove the specimen from the tester and push its broken ends together then measure its final length. 7. Convert the Dial Indicator readings into force values. 8. Subtract the Dial Indicator readings from the Load Nut movement readings to find the extension at each step. 9. Convert your force and extension values into stress and nominal strain values. 10.Plot your stress and strain results on the chart paper for your specimen (for best presentation, the steel, PVC and alloy specimens each have their own chart paper). 11.From your charts, note the yield point and tensile strength for each specimen. 12.For each specimen, find the gradient of the elastic region to compare the stiffness of the materials. Questions and Conclusions Two of the specimen materials should produce a similar gradient in the elastic region. Why is that? Why do you think a constant force application rate and temperature are so important for PVC specimens? WARNING Remember to fit the Safety Guard. NOTE For the PVC specimens, after the yield point, the Dial Indicator reading slowly drops each time you turn the Load Nut, and may continue dropping for many minutes. For this reason, you must record the value as soon as you change the load to give consistent results. Metal specimens may not break until an extension of between 1 and 5 ...

Philip MatherPhilip Mather

��

This transcript is for Philip Mather who completed a Foundation Degree Science in Applied Computing. In their first year (2013/4), they achieved grades of Pass or above in 6 modules totaling 120 credits at level 4. In their second year (2014/5), they achieved grades ranging from Pass to Merit in 6 modules totaling 120 credits at level 5. Their overall average at level 4 was 77.67% and at level 5 was 71.17%, allowing them to be awarded the Foundation Degree Science with Distinction on June 26, 2015.

IRJET- Geotechnical Investigation of Different Soil Samples using Regression ...IRJET Journal

��

This document presents the results of a study that used regression analysis to develop equations to predict geotechnical properties of soils based on index properties. Soil samples were collected from various locations and tested in the laboratory to determine their California bearing ratio (CBR), maximum dry density (MDD), and optimum moisture content (OMC). Simple and multiple linear regression analyses were performed to establish relationships between these properties and index values like liquid limit, plastic limit, plasticity index, etc. Correlation coefficients were calculated. Equations were developed for CBR, MDD and OMC in terms of percentage compositions of industrial waste, fly ash, red mud, cement and lime stabilizers. Additional equations linked CBR to individual index properties.

Presentation continuous series

1. Continuous Future Series Construction Quant Team Ruchi Agri-Trading Singapore May 03, 2013

2. Why Continuous Future Series? • We require a continuous historical price series for analysis or back-testing of trading strategies. • One difficulty with the futures historical data is associated with splicing contracts together at contract boundaries. • The results of any analysis will depend on the method we choose to construct the data series.

3. Different types of Continuous Series • Constant Month Series (S Nov1, etc.) • Constant Distance / Rolling Nearby / N-th Nearest Contract Series (S 1, BO2, etc.) • Time to maturity weighted Series (Risk Metrics linear interpolation approach) • Volume/OI weighted Series

4. Next Nearby/Rolling Nearby 13.6 13.8 14 14.2 14.4 14.6 14.8 15 15.2 Next Nearby Series S.X2 S.F3 Next.Nearby

5. Time to Maturity Weighted Series (RiskMetrics Method)

6. An Example: Time to Maturity Weighted Series Dates Contract Expiration Number of Days to Expiration Weights (%) Actual Contract Prices CMF Price NB1 1m CMF NB2 NB1 1m CMF NB2 ࣓ࡺ࡮૚ ࣓ࡺ࡮૛ ࡼࡶ࢛࢒ ࡼ࡭࢛ࢍ ࡼࡿࢋ࢖ ࡼࢉ࢓ࢌ 2011- 07-11 07-14 08-10 08-12 3 30 32 0.10 0.90 13.5250 13.4650 13.4450 13.4710 2011- 07-12 07-14 08-11 08-12 2 30 31 0.07 0.93 13.6650 13.5550 13.5500 13.5623 2011- 07-13 07-14 08-12 08-12 1 30 30 0.03 0.97 13.8700 13.7475 13.7475 13.7516 2011- 07-14 07-14 08-15 08-12 0 30 29 0.00 1.00 13.9050 13.8200 13.7975 13.8200 2011- 07-15 08-12 08-15 09-14 28 30 61 0.93 0.07 - 13.8575 13.8300 13.8557 2011- 07-18 08-12 08-17 09-14 25 30 58 0.83 0.17 - 13.8550 13.8225 13.8496

7. Time to Maturity Weighted Series 13.5 14 14.5 15 15.5 16 Constant Distance Weighted Series S.X2 S.F3 Weighted

8. Comparison - Different Methods 0 20000 40000 60000 80000 100000 120000 140000 7 8 9 10 11 12 13 14 15 16 1/8/2008 1/9/2008 1/10/2008 1/11/2008 1/12/2008 Comparion b/w Weighted/Unweighted/Generic Bloomberg Series Nearby(Actual) Weighted Generic S1 VOLUME

9. Comparison - Different Methods -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 2/1/2008 2/3/2008 2/5/2008 2/7/2008 2/9/2008 2/11/2008 Difference b/w Weighted Series & Generic BBG Series BBG Generic - Weighted

10. Comparison - Return Distributions Next Nearby Weighted Std Dev 0.018 0.016 Skewness -0.905 -0.331 Kurtosis 20.612 4.860

11. Comparison - Different Methods Method Strengths Weaknesses Nth Nearest contract No data adjustment. Represents actual values for today’s market. Gaps in historical data. Weighted Series Smoothens in data series. Good for statistical research. Non-tradable price representations. Constant Month Series Offers a long period of time in historical data on an annualized basis without adjusting the data. Small Open Interest/volume periods

�ݺ�ߣ

Presentation continuous series

Recommended

More Related Content

Similar to Presentation continuous series (20)

Presentation continuous series