Spectral variable selection based on least absolute shrinkage and selection operator with ridge-adding homotopy

Globally, technology firms are characterized by high level of innovation, rapid obsolescence of technologies, high investment risk and unpredictability of future cash flows. All these make conventional discounted cash flow valuation methods inadequate for valuation of technology firms. This study aims to develop sector regression models for relative valuation of technology firms by evaluating firm-level determinants of price multiples. Results suggest that price to book is the most appropriate multiple for valuing developed market technological firms, whereas price to sales is the most apt multiple for emerging market firms. Variable selection by least absolute shrinkage and selection operator (lasso) validates that growth rate, research intensity and cash holding influence value of price multiples for both developed market and emerging market firms. Similarly, smaller firms tend to generate higher value of the multiples under both categories. Firms’ ESG practices is an important determinant of price multiples for developed market firms, however, it does not influence the multiples’ value for emerging market firms.

Download Full-text

The successive projections algorithm for spectral variable selection in classification problems

Chemometrics and Intelligent Laboratory Systems ◽

10.1016/j.chemolab.2004.12.001 ◽

2005 ◽

Vol 78 (1-2) ◽

pp. 11-18 ◽

Cited By ~ 118

Author(s):

Márcio José Coelho Pontes ◽

Roberto Kawakami Harrop Galvão ◽

Mário César Ugulino Araújo ◽

Pablo Nogueira Teles Moreira ◽

Osmundo Dantas Pessoa Neto ◽

...

Keyword(s):

Variable Selection ◽

Classification Problems ◽

Successive Projections Algorithm ◽

Spectral Variable ◽

Successive Projections

Download Full-text

Measurement error correction in the least absolute shrinkage and selection operator model when validation data are available

Statistical Methods in Medical Research ◽

10.1177/0962280217734241 ◽

2017 ◽

Vol 28 (3) ◽

pp. 670-680 ◽

Cited By ~ 4

Author(s):

Monica M Vasquez ◽

Chengcheng Hu ◽

Denise J Roe ◽

Marilyn Halonen ◽

Stefano Guerra

Keyword(s):

Measurement Error ◽

Variable Selection ◽

Error Correction ◽

Simulated Data ◽

Correction Method ◽

Serum Biomarkers ◽

Parameter Estimates ◽

Validation Data ◽

Selection Operator ◽

Measurement Error Correction

Measurement of serum biomarkers by multiplex assays may be more variable as compared to single biomarker assays. Measurement error in these data may bias parameter estimates in regression analysis, which could mask true associations of serum biomarkers with an outcome. The Least Absolute Shrinkage and Selection Operator (LASSO) can be used for variable selection in these high-dimensional data. Furthermore, when the distribution of measurement error is assumed to be known or estimated with replication data, a simple measurement error correction method can be applied to the LASSO method. However, in practice the distribution of the measurement error is unknown and is expensive to estimate through replication both in monetary cost and need for greater amount of sample which is often limited in quantity. We adapt an existing bias correction approach by estimating the measurement error using validation data in which a subset of serum biomarkers are re-measured on a random subset of the study sample. We evaluate this method using simulated data and data from the Tucson Epidemiological Study of Airway Obstructive Disease (TESAOD). We show that the bias in parameter estimation is reduced and variable selection is improved.

Download Full-text