Overview Historical review of the FGV’s Brazilian Manufacturing Survey

Slides:



Advertisements
Similar presentations
Paul Smith Office for National Statistics
Advertisements

European Commission Directorate General Economic and Financial Affairs New question on capacity utilisation in services - state of play and way forward.
Building Up a Real Sector Confidence Index for Turkey Ece Oral Dilara Ece Türknur Hamsici CBRT.
What is Forecasting? A forecast is an estimate of what is likely to happen in the future. Forecasts are concerned with determining what the future will.
Chapter 12 - Forecasting Forecasting is important in the business decision-making process in which a current choice or decision has future implications:
Roberta Russell & Bernard W. Taylor, III
ForecastingOMS 335 Welcome to Forecasting Summer Semester 2002 Introduction.
Operations Management R. Dan Reid & Nada R. Sanders
Forecasting Revenue: An Example of Regression Model Building Setting: Possibly a large set of predictor variables used to predict future quarterly revenues.
The ECB Survey of Professional Forecasters Luca Onorante European Central Bank* (updated from A. Meyler and I.Rubene) October 2009 *The views and opinions.
Business and Management Research
Operations and Supply Chain Management
Measuring U.S. Industrial Production During a Downturn in Economic Activity Prepared for the: OECD Short-term Economic Statistics Expert Group September.
Forecasting Revenue: An Example of Regression Model Building Setting: Possibly a large set of predictor variables used to predict future quarterly revenues.
International Seminar of Early Warning and Business Cycle Indicators Moscow, November 2010 Construction of Cyclical Indicators for Ukraine on the.
Copyright 2010, The World Bank Group. All Rights Reserved. Business tendency surveys, part 2 1 Business statistics and registers.
Copyright 2010, The World Bank Group. All Rights Reserved. Business tendency surveys, part 1 1 Business statistics and registers.
Managerial Economics Demand Estimation & Forecasting.
Time series Model assessment. Tourist arrivals to NZ Period is quarterly.
Sunglasses Sales Excellence Discussion. Sunglasses Identify and describe at least one further feature of this time series data with reasons. – Sunglasses.
FORECASTING (overview)
Forecasting Parameters of a firm (input, output and products)
PROCESS Receive data from Institutes Load data in Fame database Seasonally adjust the data Calculate Composite indicators Calculate Aggregates (EU/euro.
Demand Forecasting Prof. Ravikesh Srivastava Lecture-11.
Inflation Report November Output and supply.
Time-Series Forecast Models  A time series is a sequence of evenly time-spaced data points, such as daily shipments, weekly sales, or quarterly earnings.
Main results of the Break-out session on Tendency Surveys Gian Paolo Oneto Istat.
Euro-Indicators Working Group MEASURING OUTPUT GAP IN LITHUANIA 1997–2007 Jurga Rukšėnaitė Chief Specialist, Methodology and.
Survey Training Pack Session 2 – Data Analysis Plan.
Copyright 2010, The World Bank Group. All Rights Reserved. Producer prices, part 2 Measurement issues Business Statistics and Registers 1.
Economics 173 Business Statistics Lecture 28 © Fall 2001, Professor J. Petry
United Nations Statistics Division Overview of handbook on cyclical composite indicators Expert Group Meeting on Short-Term Economic Statistics in Western.
BRAZILIAN SYSTEM OF TENDENCY SURVEYS AND CYCLE INDICATORS
Quantitative Methods for Business Studies
Chapter 5: Target Markets: Segmentation and Evaluation
THE DEMAND FOR LANGUAGE SKILLS IN THE VISEGRAD FOUR
European Commission Directorate General for Economic and Financial Affairs The harmonised EU investment survey: What can it tell us about investment growth.
How do we measure economic performance?
Quantified perceived and Expected Inflation in the Euro Area
Herman Smith United Nations Statistics Division
Forecasting Chapter 9.
Inflation Report February 2006.
Business Analysis.
Look at the following graph?
Forecasting Methods Dr. T. T. Kachwala.
Artur Andrysiak Economic Statistics Section, UNECE
Presentation by Eurostat
Service lives of R&D.
AGRICULTURE TENDENCY INDEX IN INDONESIA: PROGRESS AND CHALLENGES
Business and Management Research
Regional Workshop on Short-term Economic Indicators and Service Statistics September 2017 Chiba, Japan Alick Nyasulu SIAP.
ESCB experiences with breaks in labour market time series
Interpreting the CBI Service Sector Survey
Quality Aspects and Approaches in Business Statistics
Macroeconomic heatmap taking the temperature of the Estonian economy
Joint EU-OECD Workshop on International Development of Business and Consumer tendency Surveys Use of Business and Consumer pinion Survey Data in the ECB.
Global Assessment on Tendency Surveys
New Composite Indicators Based on KOF Business Surveys
Price Indices for External Trade of Goods Eleonora Baghy 26/05/2013
Prepared by Lee Revere and John Large
United Nations Statistics Division
LAMAS Working Group 29 June-1 July 2016
Regression Forecasting and Model Building
Chapter 8 Supplement Forecasting.
Competitive Industry Report and Calculations
Whereas chain-linking of annuals in previous years prices is unambiguous, it is not at quarterly frequencies. Contrary to the US, annuals as well as quarters.
BEC 30325: MANAGERIAL ECONOMICS
Business Analysis.
Short Term Statistics in National Accounts
Issues on Seasonal Adjustment in the EECCA countries
Presentation transcript:

Overview Historical review of the FGV’s Brazilian Manufacturing Survey The challenge of turning it into a monthly Survey Detecting seasonality in the historical quarterly series Choosing a method for adjusting seasonality Analysing potential effects on seasonality while migrating the frequency from quarterly to monthly Concluding Remarks

Historical Review Brazilian Quarterly Manufacturing Survey created in 1966; Inspired by european surveys (INSEE, Ifo); 1,100 responses each month; Turned into monthly frequency in Nov. 05.

Challenge of moving into the Monthly Frequency Potential problems: Complexity of the questionnaire – Some questions are related to the firm’s specific lines of products, not just the company as a whole; Respondent Fatigue - Companies complain about the time spent answering surveys in Brazil; No enforcement - Companies are not obliged to answer the questionnaires.

Challenge of moving into the Monthly Frequency Potential Solutions: Dividing the sample into two different panels of companies with similar profiles; Applying the same questionnaire to all respondents in all editions; Chosen Movement: Applying to all respondents the same questionnaire in the original months; Applying a smaller questionnaire to all respondents in the complementary editions.

Apparent Reasons for Seasonality in the Brazilian Manufacturing Survey From 14 series of the survey analysed: 8 (57.1%) presented pronounced seasonal pattern 6 (42.9%) did not present a pronounced seasonal pattern There are apparently two reasons for the presence of seasonality in the survey series: The profile of the variable being measured The form of presenting the question (phrasing)

Reasons for Seasonality Type of question Month of the Survey Previous Quarter Following Quarter Actual Results Forecasts Examples: Exception: Future Production Future Employment Future Prices Furture Business Situation (compared to same semester, last year)

Reasons for Seasonality Type of variable Example: Level of Capacity Utilisation (reference to the month of the survey, multiple choice) 0% 1% to 20% ... 80% to 89% 90% to 99% 100% 9 options: Other Example: Exception: Level of Demand Level of Stocks Present Business Situation (slight seasonal pattern)

Future Production Indicator * Strong seasonal pattern Quarterly Data from Jan.05 to Jul.09 * Indicator = Balance + 100

Level of Stocks Indicator * Slight or no seasonal pattern Quarterly Data from Jan.05 to Jul.09 * Indicator = Balance + 100

Seasonally adjusting the monthly series For building a monthly Confidence Indicator with the most relevant indicators of the Survey, FGV had to seasonally adjust the monthly series; In 2008, with just three years of monthly series we decided to test the interpolation of the quarterly series using Kalman filters in a structural model framework; The results were considered to be a success: Seasonally adjusted monthly data of the industry tendency survey gained more relevance as a reference to the Brazilian business cycle; Even with the short monthly time series available, and using an univariate interpolation method, the series appear consistent and present a good fit when compared to quantitative indicators.

Adjusting the short time monthly series

Testing Interpolation

LCU – Original and Adjusted

Choosing the Interpolation Method

Evaluating Seasonality after four years of Monthly Frequency (Nov. 05)

Level of Capacity Utilisation Comments: Factors change along time but have stabilised in the ’00 decade, specially after 2002

Future Production Comments: Seasonal factors are continuously changing but there are no signs of strutuctural changes around 2005

Level of Demand Comments: No changes across time

Future Employment Comments: Changes occur along time. After 2005 there seems to be no structural break but since 2004, the factors for the month of July (maximum) started to increase (03 = 4.2; 04 = 4.4; 05 = 4.7; 09 = 6.1)

Future Employment

Industrial Production

Concluding Remarks Questionnaire phrasing makes a difference. Seasonality appears more pronounced in questions that imply some kind of comparison over time; After four years, there is no sign that the collection of data in the other months of the year, has changed the relative seasonality pattern of the original months; In 2008-2010 FGV is creating Services, Commerce and Building Surveys, containing question phrasing that intend to correct seasonality from the start. In a few years we will be able to analyse whether this measure was successful in breaking or reducing seasonality patterns.

Thank You !