Department seminar speaker

Events

Departmental and Data Science seminar series

Termly events held by the Department of Methodology and the Social and Economic Data Science (SEDS) research unit.

 

Leading social scientists consider cutting-edge quantitative and qualitative methodologies, analyse the logic underpinning an array of approaches to empirical enquiry, and discuss the practicalities of carrying out research in a variety of different contexts.

Department of Methodology Seminar Series 

Department seminars take place in COL 8.13 from 12:30 - 14:00 unless otherwise stated. A sandwich lunch will be provided from 12:15.


Kuha3

Estimating regression models with latent variables: One, three, or two steps?
Speaker: Jouni Kuha, Department of Methodology/Department of Statistics, LSE
Date: 25th January 2018
Abstract: We consider statistical models which combine measurement models for latent variables with regression models (“structural models”) where the latent variables are explanatory or response variables. Such models can be estimated in a number of different ways. One-step estimation estimates all parts of the model at the same time.  This has the disadvantage that changes in the specification of the structural model also change the estimated measurement model and thus in effect change the definition of the latent variables. Stepwise methods avoid this problem by separating the estimation of the measurement and structural models. In three-step estimation, the first step is to  estimate the measurement model alone, the second to use this model to assign values for the latent variables, and the third to use these assigned values to estimate the structural model. We propose instead a two-step approach which carries only the estimated parameters forward from the first step and thus avoids the ultimately unnecessary step of assigning values for the latent variables. We present applied examples to illustrate the methods, and simulation studies to demonstrate the behaviour of the estimators. 

StephenKhan

Beyond academia: communicating research to broader audiences
Speaker: Stephen Khan, Editor of The Conversation, UK
Date: 1st February 2018
Abstract: In a time of massive flows of data and information, making new academic research relevant and useful has become a more tangled task. It is no longer a matter of only publishing and disseminating findings in the "top" journal or academic conference. It is also a matter of finding alternative "avenues" or streams of information that intersect with the individuals and communities who have a word in the phenomenon or problem in question. Reaching these audiences is critical to secure that the rigorous work done by researchers meets with the political and social mechanisms that enable knowledge to transform reality.
Identifying and navigating these alternative streams of information pose new challenges to the modern researcher. To discuss and address some of them, the Research and Methods Society at LSE has invited Stephen Khan –editor of The Conversation in the UK– to share their experience on communicating academic research through online journalism. The talk will focus on how to reach non-specialized audiences, as well as on the difficulties of rebuilding the public's trust in research.

Stephen Khan is The Conversation’s Editor in the UK. He was a news editor at The Guardian and previously Deputy Foreign Editor of The Independent, Scotland Editor of The Observer and also worked for The Sunday Herald in Scotland.

Chris-Chaplin

A Movement of Fragments: An ethnographic analysis of Indonesian Salafi Islam
Speaker: 
Chris Chaplin, Department of Methodology, LSE
Date: 8th February 2018
Abstract: The spread of Salafi Islam across Indonesia has accelerated apace ever since its introduction into the archipelagic nation during the mid-1980s. Propagating a ‘literalist’ interpretation of Islam that is closely affiliated to scholars and Islamic institutions in the Arabian Peninsula, Salafi activists place strong emphasis on the need to separate themselves from society by dividing the world into those who follow ‘true’ Islam and those who do not. This is not without controversy, as such religious boundary making has led Salafis to implement strict gender segregation, set up their own enclaves, and differentiate themselves through dress and practice from more established Islamic traditions within Indonesia. Yet, such boundary implementation is rarely clear-cut, but instead rife with disagreement, negotiation and tension. Drawing from his ethnographic experience with urban Salafi activists in Yogyakarta as well as the broader idea of ‘working epistemics’ (Voyer and Tondman 2017), Chris will examine how the inter-subjective encounters that are the core of fieldwork shed light onto the ways Islamic tenets are challenged, implemented and refined on a daily basis. Investigating how activists propagate the movement, conduct themselves in private as well as engage with their local surroundings, Chris argues that the implementation of socio-religious boundaries remain open to contextual considerations and negotiation. They are based as much, if not more, on horizontal relationships between activists as they are on any belief in a universal Salafi truth. This has significant implications as to how we must understand Salafism. Instead of being a coherent movement promoting a ‘timeless’ religious truth, Salafism is a multi-layered movement of disparate networks, prone to rupture, disagreement and contextual adaptation. 

1 March

Exact details TBC

 

Social and Economic Data Science Seminars

Data Science seminars are run in conjunction with the Departments of Mathematics and Statistics. These seminars take place from 16:15 - 17:45 in COL 8.13 (unless otherwise stated).

DavidParkes-profile

Optimal Economic Design through Deep Learning
Speaker:
David C. Parkes, Paulson School of Engineering and Applied Sciences Harvard University
Date:
17 January 2018
Abstract: Designing an auction that maximizes expected revenue is an intricate task. Despite major efforts, only the single-item case is fully understood. We explore the use of tools from deep learning on this topic. The design objective that we adopt is revenue optimal, dominant-strategy incentive compatible auctions. For a baseline, we show that multi-layer neural networks can learn almost-optimal auctions for a variety of settings for which there are analytical solutions, and even without leveraging characterization results. We also show that deep learning can be used to derive auctions for poorly understood problems, including settings with multiple items and budget constraints. Our research also demonstrates that the deep learning framework is quite general, being applicable to other problems of optimal economic design.

Joint work with Paul Duetting (LSE), Zhe Feng (Harvard University), and Harikrishna Narasimhan (Harvard University). Working paper: https://arxiv.org/abs/1706.03459

MicroMusolesi

Modelling Human Behaviour using Mobile Data
Speaker: Mirco Musolesi
Date: 1st February
Abstract: We constantly generate digital traces in our online and offline lives, for example by using our smartphones, by interacting with everyday devices and the technological infrastructure of our cities or simply by posting content on online social media platforms. This information can be used to model and possibly predict human behaviour in real-time, at a scale and granularity that were unthinkable just a few years ago. 
In this talk, Mirco will present recent work in modelling human behaviour using these "digital traces” with a specific focus on mobile data. He will provide an overview of the methodological, algorithmic, and systems issues related to the development of solutions that rely on the online analysis and modelling of this type of data. As a case study, he will show how mobile phones can be used to collect and analyse mobility patterns of individuals in order to quantitatively understand how mental health problems affect their daily routines and behaviour and how potential changes can be automatically detected. Hr will demonstrate that it is possible to observe a non trivial correlation between mobility patterns and depressive mood using data collected by means of smartphones. Finally, he will also introduce his and his researchers' efforts in using cellular data for modelling mobility patterns of individuals at scale and their applications in the area of data for development.

Roel

Electronic FX trading – where Game Theory meets Data Science
Speaker: Roel Oomen, Global co-head of electronic FX spot trading, Deutsche Bank, London
Date: 22nd February 2018
Abstract: In this talk, Roel will discuss recent developments in electronic FX trading and show how data science applied to dense (as opposed to big) financial data can be used to make practical decisions around execution optimisation. He will go over a  number of case studies taken from a live trading environment.

Omar Guerrero

How do governments determine policy priorities? Studying development strategies through networked spillovers.
Speaker: Omar Guerrero, Said Business School, University of Oxford
Date: 8th March 2018
Abstract: Determining policy priorities is a challenging task for any government. The interdependency between policies and corruption of government officials creates a rugged landscape that governments need to navigate in order to reach their goals. We develop a framework to model the evolution of development indicators as a public goods game on a network. Our approach accounts for the complex network of interactions among policy issues as well as the principal-agent problem arising from budget assignment. Using development indicator data from more than 100 countries over 11 years, our main results are as follows: (i) well known empirical patterns involving aggregate corruption and income can be explained by the opaque relationship between policy outcomes and contributions of public agencies; (ii) achieving a multidimensional target depends on a learning process during the allocation of resources; (iii) the network of spillover effects provides country-specific context that is critical to order policy priorities; and (iv) a country may reach different development targets but how `easy' it is and through which policies it can be achieved may vary considerably. Our framework provides an analytic tool to generate bespoke advise on development strategies.