Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
The Power of the Synthetic Control Method
Dalarna University, School of Technology and Business Studies, Microdata Analysis.ORCID iD: 0000-0003-2317-9157
Dalarna University, School of Technology and Business Studies, Microdata Analysis.
2016 (English)Report (Other academic)
Abstract [en]

The synthetic control method (SCM) is a new, popular method developed for the purpose of estimating the effect of an intervention when only one single unit has been exposed. Other similar, unexposed units are combined into a synthetic control unit intended to mimic the evolution in the exposed unit, had it not been subject to exposure. As the inference relies on only a single observational unit, the statistical inferential issue is a challenge. In this paper, we examine the statistical properties of the estimator, study a number of features potentially yielding uncertainty in the estimator, discuss the rationale for statistical inference in relation to SCM, and provide a Web-app for researchers to aid in their decision of whether SCM is powerful for a specific case study. We conclude that SCM is powerful with a limited number of controls in the donor pool and a fairly short pre-intervention time period. This holds as long as the parameter of interest is a parametric specification of the intervention effect, and the duration of post-intervention period is reasonably long, and the fit of the synthetic control unit to the exposed unit in the pre-intervention period is good.

Place, publisher, year, edition, pages
Borlänge: Högskolan Dalarna, 2016. , p. 15
Series
Working papers in transport, tourism, information technology and microdata analysis, ISSN 1650-5581 ; 2016:10
Keywords [en]
Bootstrap; Comparative case study; Counterfactual analysis; Intervention effect; Monte Carlo Simulation; Statistical Inference
National Category
Probability Theory and Statistics
Research subject
Complex Systems – Microdata Analysis, General Microdata Analysis - methods
Identifiers
URN: urn:nbn:se:du-23901OAI: oai:DiVA.org:du-23901DiVA, id: diva2:1061375
Funder
Swedish Retail and Wholesale Development CouncilAvailable from: 2017-01-02 Created: 2017-01-02 Last updated: 2018-08-21Bibliographically approved
In thesis
1. Who benefits when IKEA enters local markets in Sweden?: An empirical assessment using difference-in-difference analysis, synthetic control methods, and Twitter sentiment analysis
Open this publication in new window or tab >>Who benefits when IKEA enters local markets in Sweden?: An empirical assessment using difference-in-difference analysis, synthetic control methods, and Twitter sentiment analysis
2018 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Policy makers often spend considerable amounts of money to attract IKEA to their region despite not having any empirical measurements on its expected contribution to the local economy. As such, an empirical study of the economic and social impact of new IKEA stores can aid political decision making, and contribute to the literature regarding how big-box retail entry affects the regions where they enter.

This dissertation aims to estimate: the impact of IKEA entry on incumbent retailers productivity, and investigate if the impact is heterogenus depending on local maket size, type of retail industry, distance to surrounding retailers, and firm size; IKEA entry effects on the average labor productivity in durable goods retailing in the entry regions; and, finally, public opinions regarding  IKEA entry.

For IKEA entry effects on incumbent retailers, Paper I~III separately examine four factors of potential heterogeneity. Paper I finds that market size matters: smaller rural regions have bigger IKEA effects. Paper II considers two factors: firm industry and distance, and confirms that IKEA entry effects dissipate over distance. The positive impact of IKEA entry on incumbent retailers is limited to those selling complementary goods to IKEA. No positive effects were found for the urban entry in Gothenburg in the two first papers, which is somewhat surprising. Paper III found that a positive effect exist also in Gothenburg, but it is limited to relatively small incumbent retailers with a capital stock below 1 500 000 SEK. Policy making tends to consider IKEA overall effects on entry municipalities besides IKEA spillover effects on firms. Paper V shows that rural regions are affected by IKEA entry, while larger urban markets are not.

For the social effects of IKEA, Paper VI uses Twitter text mining to study public opinions regarding IKEA entry into local markets. The new IKEA stores under study caught significant public attention at the time of entry, with mostly positive attitudes toward the new stores. The favorite topics for discussion at the time of the different IKEA entries were heterogeneous depending on location.

Methodologically, Paper I uses traditional Difference-in-Difference (DID) to have an initial understanding of IKEA entry spillover effects in four regions; Paper II extends to Spatial DID to catch the spatial interaction between firms; Paper III uses Panel Smooth Transition Regression to identify heterogenous effects due to firms size. Paper IV investigates a new treatment effects estimation aproach, Synthetic Control Method (SCM), to explore when the SCM is powerful, and how to improve its performance; Paper V then uses SCM to estimate IKEA effects at municipality level. In addition, to make SCM developed readily available for other researchers, the author of this thesis also published one web-application to implement a synthetic control method power test, and another to implement parametric & non-parametric estimation and inference.  

These findings confirm that IKEA has a positive effect on the regions where they enter. Nevertheless, governments that are to decide if to allow a big-box retail entry into their local community should be aware that the impact of such entry will depend on the size of the existing retail market, the type of existing retail industry, and the size of existing retailers in the entry region.

Place, publisher, year, edition, pages
Borlänge: Högskolan Dalarna, 2018
Series
Dalarna Doctoral Dissertations ; 7
Keywords
big-box entry effect; spatial spill-over effect; heterogeneous causal effect; difference-in-difference; synthetic control method; sentiment analysis.
National Category
Economics and Business
Research subject
Complex Systems – Microdata Analysis
Identifiers
urn:nbn:se:du-27896 (URN)978-91-85941-85-8 (ISBN)
Public defence
2018-09-14, Clas Ohlson, Borlänge, 13:00 (English)
Opponent
Supervisors
Funder
Swedish Retail and Wholesale Development Council, 2015:4
Available from: 2018-08-21 Created: 2018-06-14 Last updated: 2018-08-21Bibliographically approved

Open Access in DiVA

fulltext(637 kB)201 downloads
File information
File name FULLTEXT01.pdfFile size 637 kBChecksum SHA-512
0ce6f6ce5ed341ba3d412085b541ca1fac92194494f2ec10b118fd245a18b6fb635d7f32ad8bce4948cfd77b1c1b986355bb1f3627ac5a8003099251e35c705e
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Carling, KennethLi, Yujiao
By organisation
Microdata Analysis
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar
Total: 201 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 1142 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf