Sandro Leidi
|
Sandro worked for the University of Reading, UK for over 20 years, providing consultancy and CPD courses for both the public and private sectors. During this time he accumulated a wealth of experience in many practical applications of statistical modelling.
He is an experienced user of both linear and generalised linear mixed models for complex designs such as cross-overs with repeated measures. In particular, he specialises in the analysis of discrete responses: counts, proportions and ordered categories (from surveys of customers' degree of satisfaction for example), as well as modelling of multivariate responses which are clustered but measure different attributes, hence require separate supporting distributions. He also conducts joint modelling of longitudinal (repeated measures) data and time-to-event outcomes. He often runs causal models that use the potential outcomes framework to estimate the average treatment effect in observational studies. Sandro also applies survival analysis (time-to-event) techniques, such as the Cox and Weibull models with frailties, and is experienced in using the counting process format for modelling with time-varying covariates, as well as models for competing risks. As for missing data, he can conduct multiple imputation by chained equations on both multilevel data and longitudinal data. He is experienced at conducting power studies to derive the required sample size for studies with multilevel clustered data, in both cross-sectional and longitudinal structures. He has an excellent knowledge of mainstream statistics packages such as Genstat, Minitab, R, SAS, SPSS and Stata, as well as specialised packages for power studies such as PASS. As for the Bayesian flavour of modelling, he uses WinBUGS, OpenBUGS, JAGS from within R, and the bayesian facilities in SAS, SPSS and Stata. Notably, he is a power user of Excel, versed in the use of array functions. For a personally tailored consultancy discussion, feel free to contact Sandro in English, French, Spanish, Italian or Catalan. Latest publications: 2021: A small area estimation method for investigating the relationship between public perception of drug problems with neighborhood prognostics: trends in London between 2012 and 2019. International Journal of Environmental Research and Public Health. 2021: Multiple morbidities in an inner-city English substance misuse treatment service: hierarchical cluster analysis to derive treatment segments. Journal of Dual Diagnosis. 2021: Estimating a treatment effect on reoffending by people in prison with an alcohol use disorder: a national matched propensity score analysis. Journal of Offender Rehabilitation. 2020: Estimating a treatment effect on recidivism for correctional multiple component treatment for people in prison with an alcohol use disorder in England. Substance abuse treatment, prevention and policy. 2019: Studying biological sciences does not lead to adoption of a healthy lifestyle. Perspectives in Public Health. 2019: Factors associated with concurrent illicit use of opiates and crack/cocaine among opiate-users in treatment: Implications for treatment services in England. Addiction Research & Theory. 2017: Clinical outcomes and cost-effectiveness of brief guided parent-delivered cognitive behavioural therapy and solution-focused brief therapy for treatment of childhood anxiety disorders: a randomised controlled trial. THE LANCET Psychiatry. Recent Career History
|