Université de Fribourg

Machine learning with screens for detecting bid-rigging cartels

Huber, Martin ; Imhof, David

In: International Journal of Industrial Organization, 2019, vol. 65, p. 277-301

We combine machine learning techniques with statistical screens computed from the distribution of bids in tenders within the Swiss construction sector to predict collusion through bid-rigging cartels. We assess the out of sample performance of this approach and find it to correctly classify more than 84% of the total of bidding processes as collusive or non-collusive. We also discuss...

Université de Fribourg

Transnational machine learning with screens for flagging bid-rigging cartels

Huber, Martin ; Imhof, David ; Ishii, Rieko

(Working Papers SES ; 519)

We investigate the transnational transferability of statistical screening methods originally developed using Swiss data for detecting bid-rigging cartels in Japan. We find that combining screens for the distribution of bids in tenders with machine learning to classify collusive vs. competitive tenders entails a correct classification rate of 88% to 93% when training and testing the method...

Université de Fribourg

Machine learning approach for flagging incomplete bid-rigging cartels

Wallimann, Hannes ; Imhof, David ; Huber, Martin

(Working Papers SES ; 513)

We propose a new method for flagging bid rigging, which is particularly useful for detecting incomplete bid-rigging cartels. Our approach combines screens, i.e. statistics derived from the distribution of bids in a tender, with machine learning to predict the probability of collusion. As a methodological innovation, we calculate such screens for all possible subgroups of three or four bids...

Université de Fribourg

Empirical Methods for Detecting Bid-rigging Cartels

Imhof, David ; At, Christian (Dir.) ; Madiès, Thierry (Dir.)

Thèse de doctorat : Université de Fribourg, 2018.

Organisée en cinq chapitres, cette thèse de doctorat vise à étudier et à développer des méthodes empiriques pour détecter les cartels de soumission. Elle en propose également une étude économique et étudie leur fonctionnement notamment en se fondant sur l’expérience professionnelle de l’auteur. Chaque chapitre correspond à un article indépendant s’intégrant de façon...

Université de Fribourg

Machine learning with screens for detecting bid-rigging cartels

Huber, Martin ; Imhof, David

(Working Papers SES ; 494)

We combine machine learning techniques with statistical screens computed from the distribution of bids in tenders within the Swiss construction sector to predict collusion through bid-rigging cartels. We assess the out of sample performance of this approach and find it to correctly classify more than 80% of the total of bidding processes as collusive or non-collusive. As the correct...

Université de Fribourg

Simple Statistical Screens to Detect Bid Rigging

Imhof, David

(Working Papers SES ; 484)

The paper applies simple statistical screens to a bid-rigging cartel in Switzerland, and shows how well the screens detect it by capturing the impact of collusion on the discrete distribution of the bids. In case of bid rigging, the support for the distribution of the bids decreases involving a lower variance, illustrated by the coefficient of variance and the kurtosis statistic. Furthermore,...

Université de Fribourg

Econometric tests to detect bid-rigging cartels: does it work?

Imhof, David

(Working Papers SES ; 483)

This paper tests how well the method proposed by Bajari and Ye (2003) performs to detect bidrigging cartels. In the case investigated in this paper, the bid-rigging cartel rigged all contracts during the collusive period, and all firms participated to the bid-rigging cartel. The two econometric tests constructed by Bajari and Ye (2003) produce a high number of false negative results: the tests do...

Université de Fribourg

Screening for bid-rigging - does it work?

Imhof, David ; Karagök, Yavuz ; Rutz, Samuel

(Working Papers SES ; 468)

This paper proposes a method to detect bid-rigging by applying mutually reinforcing screens to a road construction procurement data set from Switzerland in which no prior information about collusion was available. The screening method is particularly suited to deal with the problem of partial collusion, i.e. collusion which does not involve all firms and/or all contracts in a specific data set,...