The thesis is framed around three studies that can be seen as testing conjectures about whether contextual factors, which are not part of standard economics models, are sufficiently important determinants of behaviour that they should be included in those models. The three studies also cast some light -on the important ~ but poorly understood - issue of the contextual interplay between individual preference and (non)strategic decision-making. They do so by focussing on market valuations, price-judgement tasks with anchoring exposure and coordination games, respectively. The first study reports a new demonstration that participants in experimental markets adjust their bids towards the price observed in previous market periods when - by design - individuals' values should not be affiliated with the market price. The second study presents new experimental evidence on the effects of arbitrary anchors on economic valuations that fails to replicate the results of an influential study with far reaching economic implications. Through the lens of the provided model, the study offers also insights into the optimal usage of experimental methods and the appropriate use of experimental results. The third study, after introducing a novel technique (due to Cialdini et al., 1997) for measuring 'oneness' (or, in other words, the closeness of the social relations that exist within groups of players), reports the results of experiments investigating the influence of social relations in a weak-link game. While investigating the relevance of 'context' is a connecting theme as well as resorting to experimentation is a common feature of the thesis, the three studies offer also relevant methodological contributions ranging from test-driving psychological tools as far as providing insights into the mechanics of inference and into the function of the economics knowledge system.

The first chapter of this thesis gives a general discussion of the issues experienced when standard tests are applied to an economic time series when either or both of the order of integration of a series or the distribution driving the innovation sequence is unknown. In the second chapter a new procedure for detecting additive outliers in a univariate time series is proposed based on a bootstrap implementation of the test of Pen'on and Rodr'lguez (2003, Journal of Time Series Analysis 24, 193-220). This procedure is used to test the null hypothesis that a time series is uncontaminated by additive outliers against the alternative that one or more additive outliers are present. In the third chapter a testing procedure for the presence of a deterministic linear trend in a univariate time series is developed which is robust to whether a series is 1(0) or 1(1) and requires no knowledge of the form of weak dependence in the data. In the fourth chapter a test for the presence of nonlinear deterministic components in a univariate time series is developed, where the nonlinear deterministic component of a series is approximated using a Fourier series expansion, that is designed to be asymptotically robust to the order of integration of the process and to any weak dependence present in the data. In the fifth chapter tests for the presence of deterministic components in a seasonally observed univariate time series are developed . These tests are designed to be asymptotically robust to the order of integration of the series at both the zero and seasonal frequencies. In the sixth and final chapter concluding remarks are made about the results from all other chapters of the thesis, and avenues for further research are detailed. "

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.