Raju Rimal

• Rimal, Raju; Almøy, Trygve & Sæbø, Solve (2020). Comparison of multi-response estimation methods. Chemometrics and Intelligent Laboratory Systems. ISSN 0169-7439. 205. doi: 10.1016/j.chemolab.2020.104093. Full text in Research Archive
• Rimal, Raju; Almøy, Trygve & Sæbø, Solve (2019). Comparison of multi-response prediction methods. Chemometrics and Intelligent Laboratory Systems. ISSN 0169-7439. 190, p. 10–21. doi: 10.1016/j.chemolab.2019.05.004. Full text in Research Archive
• Helland, Inge Svein; Sæbø, Solve; Almøy, Trygve & Rimal, Raju (2018). Model and estimators for partial least squares regression. Journal of Chemometrics. ISSN 0886-9383. 32(9). doi: 10.1002/cem.3044. Show summary

  Partial least squares (PLS) regression has been a very popular method for prediction. The method can in a natural way be connected to a statistical model, which now has been extended and further developed in terms of an envelope model. Concentrating on the univariate case, several estimators of the regression vector in this model are defined, including the ordinary PLS estimator, the maximum likelihood envelope estimator, and a recently proposed Bayes PLS estimator. These are compared with respect to prediction error by systematic simulations. The simulations indicate that Bayes PLS performs well compared with the other methods.

• Rimal, Raju; Almøy, Trygve & Sæbø, Solve (2018). A tool for simulating multi-response linear model data. Chemometrics and Intelligent Laboratory Systems. ISSN 0169-7439. 176, p. 1–10. doi: 10.1016/j.chemolab.2018.02.009. Full text in Research Archive Show summary

  Data science is generating enormous amounts of data, and new and advanced analytical methods are constantly being developed to cope with the challenge of extracting information from such “big-data”. Researchers often use simulated data to assess and document the properties of these new methods, and in this paper we present an extension to the R-package simrel, which is a versatile and transparent tool for simulating linear model data with an extensive range of adjustable properties. The method is based on the concept of relevant components, and is equivalent to the newly developed envelope model. It is a multi-response extension of R-package simrel which is available in R-package repository CRAN, and as simrel the new approach is essentially based on random rotations of latent relevant components to obtain a predictor matrix X, but in addition we introduce random rotations of latent components spanning a response space in order to obtain a multivariate response matrix Y. The properties of the linear relation between X and Y are defined by a small set of input parameters which allow versatile and adjustable simulations. Sub-space rotations also allow for generating data suitable for testing variable selection methods in multi-response settings. The method is implemented as an update to the R-package simrel.

View all works in Cristin

• Rimal, Raju (2021). Long-term Trend on Tumour Thickness of Melanoma in Norway. Show summary

  Introduction: Norway ranks fifth in incidence and second in mortality of cutaneous melanoma worldwide. Tumour thickness at diagnosis is the cornerstone of melanoma classification and the most important prognostic factor for clinically localized primary melanoma. Increased incidence of thin tumours may be a result of increased awareness and changes in pathological practice. Recently digitized tumour thickness data (1980-2007) and data from the Melanoma Registry (2008-), enable investigating long-term trends in melanoma thickness. Aim: To investigate trends in tumour thickness, overall and in important subgroups in a nationwide case series from 1980 to 2019. Methods: Tumour thickness was categorized in T-categories: T1 (≤1.0 mm), T2 (1.0-2.0 mm), T3 (>2.0-4.0 mm), and T4 (>4.0 mm). Incidence rates were age-standardized using the European standard population. Trends and changes in incidence rate over time were analysed with annual percentage changes (APC) and average APC. Conclusions: T1 melanomas had the largest increase in incidence over time. An increasing trend was also observed in thicker tumours, suggesting that the rise in melanoma incidence is due to not only overdiagnosis/pathological practice.

• Rimal, Raju; Robsahm, Trude Eid; Green, Adèle C.; Ghiasvand, Reza; Rueegg, Corina Silvia & Bassarova, Assia Vassileva [Show all 11 contributors for this article] (2021). Long-term trends in melanoma tumour thickness in Norway.
• Rimal, Raju; Almøy, Trygve & Sæbø, Solve (2019). A prediction comparison of some multivariate methods. Show summary

  A linear model is a widely used relationship structure which we encounter. A versatile tool to simulate linear model data controlling various aspects of it can be useful not only for comp aring methods, algorithms and models but also accessing and understanding their properties. Here we will present an R-package `simrel` that can control properties such as collinearity bet ween the variables, information content in predictors that is relevant for responses and position of the predictor components that contain this information. With few parameters, the packa ge can simulate data with a wide range of properties. Various multivariate prediction methods are developed over time to address the problem related to such properties. In the second part of the talk, we will compare relatively established methods such as Principal Component Regression, Partial Least Squares Regression together with newly developed envelope methods using the multi-response data simulated with varying properties. These methods deal with the relevant and irrelevant regression structure in a different way. This part will pr esent these differences and make some comparisons on these methods.

• Rimal, Raju; Almøy, Trygve & Sæbø, Solve (2018). A versatile tool for simulating linear model data. Show summary

  Data science is generating enormous amounts of data, and new and advanced analytical methods are constantly being developed to cope with the challenge of extracting information from such “big-data”. Researchers often use simulated data to assess and document the properties of these new methods, and in this paper, we present an extension to the R-package simrel, which is a versatile and transparent tool for simulating linear model data with an extensive range of adjustable properties. The method is based on the concept of relevant components and is equivalent to the newly developed envelope model. It is a multi-response extension of R-package simrel which is available in R-package repository CRAN, and as simrel the new approach is essentially based on random rotations of latent relevant components to obtain a predictor matrix X, but in addition we introduce random rotations of latent components spanning a response space in order to obtain a multivariate response matrix Y. The properties of the linear relation between X and Y are defined by a small set of input parameters which allow versatile and adjustable simulations. The method is implemented as an update to the R-package simrel.

• Rimal, Raju (2016). A comparetive study on PCR, PLS, Envelope and BayesPLS models.
• Rimal, Raju (2019). Exploration of Multi-response Multivariate Methods. Norwegian University of Life Sciences. ISSN 978-82-575-1636-9. Show summary

  En lineær regresjonsmodell definerer et lineært forhold mellom to eller flere tilfeldige variabler. De tilfeldige variablene som er avhengige av andre tilfeldige variabler, kalles ofte responsvariabler, og de uavhengige tilfeldige variablene kalles prediktorvariabler. I de fleste tilfeller er ikke all variasjon relevant for regresjon, dvs. bare en viss mengde variasjonen i prediktorene er relevante, og bare for en del av variasjonen i responsen. Dette fører til en reduksjon av den lineære regresjonsmodellen der man kan forestille seg et underrom av rommet som spennesut av prediktorvariablene som inneholder all relevant informasjon for et underrom av rommet spent ut av responsvariablene. I denne avhandlingenprøver vi å sammenligne noen nye metoder som er basert på Envelopemodellen og noen etablerte metoder som principal komponent regresjon (PCR) og partiell minste kvadraters regresjon (PLS). Sammenligningen tester disse metodene på deres ytelse til å produsere minimum prediksjon- og estimeringsfeil, mens modelleringsdata simuleres med spesielt designede egenskaper. For simuleringen har vi også laget en R-pakke kalt simrel med et webgrensesnitt. En simuleringsmodell for multirespons, multivariat lineær modell, som simuleringsverktøyet bygger på, diskuteres i den første artikkelen. Denne artikkelen utarbeider et grunnleggende fundament for simuleringene ba- sert på konseptet om reduksjon av regresjonsmodeller. Den andre artikke- len diskuterer likhetene i Envelope-, PCR- og PLS-populasjonsmodellene. Denne artikkelen sammenligner prediksjonsytelsen til flere multivariate metoder ved bruk av en modell med en enkelt respons. De to siste artiklene gir en grundig evaluering av prediksjons- og esti- meringsegenskapene til etablerte metoder (PCR, PLS1 og PLS2) og ny- utviklede envelope-baserte metoder (Xenv og Senv). Ikke uventet fant stu- dien at det ikke finnes en enkelt metode som dominerer i alle situasjoner, men resultatene deres avhenger av egenskapene til dataene de modellerer. Imidlertid har envelope-baserte metoder vist bemerkelsesverdig resultater i mange tilfeller, både når det gjelder prediksjon og estimering. Studien anbefaler også forskere å bruke og evaluere envelope-metodene.

View all works in Cristin