Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Hur estimeras riktkurser av analytikerna? Analys av estimerade aktiepriser på Stockholmsbörsen från år 2000 – 2024 med maskininlärningsmodellen Random Forest

Mikler, Carl-Douglas LU (2026) NEKH03 20252
Department of Economics
Abstract
Stock analysts have a great influence on how investors choose to invest their money and
which companies that will receive a share of the ever-growing amount of money that Swedish
households choose to invest in stocks. With the influence that analysts have on the company
inflows and outflows of capital, one can wonder what drives their target prices.
This report analyzes which financial variables most affect quarterly consensus estimated stock
prices for companies on the Stockholm Stock Exchange from 2000 to 2024, by using the
machine learning model Random Forest. The main part of the report describes the approach of
how to use machine learning to obtain a result that shows which variable is most important,
this is described... (More)
Stock analysts have a great influence on how investors choose to invest their money and
which companies that will receive a share of the ever-growing amount of money that Swedish
households choose to invest in stocks. With the influence that analysts have on the company
inflows and outflows of capital, one can wonder what drives their target prices.
This report analyzes which financial variables most affect quarterly consensus estimated stock
prices for companies on the Stockholm Stock Exchange from 2000 to 2024, by using the
machine learning model Random Forest. The main part of the report describes the approach of
how to use machine learning to obtain a result that shows which variable is most important,
this is described step by step.
Extensive acquisition and formatting of financial variables through S&P Capital IQ Pro was
significantly time-consuming due to limitations in the database. Hyperparameter settings for
missForest and Random Forest were used before they were applied to the final testing data.
Based on the results, technical variables such as price movements were most important to the
analysts, but also variables connected to fundamental analysis.
The conclusion was not particularly surprising. It became a new insight that the technical
analysis would be so dominant as it was.
The work has provided many new insights and lessons when it comes to machine learning,
programming and data management. (Less)
Please use this url to cite or link to this publication:
author
Mikler, Carl-Douglas LU
supervisor
organization
course
NEKH03 20252
year
type
M2 - Bachelor Degree
subject
keywords
Stockholm Stock Exchange, Target prices, Random Forest, Machinelearning, R-script
language
Swedish
id
9220232
date added to LUP
2026-02-04 08:25:30
date last changed
2026-02-04 08:25:30
@misc{9220232,
  abstract     = {{Stock analysts have a great influence on how investors choose to invest their money and 
which companies that will receive a share of the ever-growing amount of money that Swedish 
households choose to invest in stocks. With the influence that analysts have on the company 
inflows and outflows of capital, one can wonder what drives their target prices. 
This report analyzes which financial variables most affect quarterly consensus estimated stock 
prices for companies on the Stockholm Stock Exchange from 2000 to 2024, by using the 
machine learning model Random Forest. The main part of the report describes the approach of 
how to use machine learning to obtain a result that shows which variable is most important, 
this is described step by step. 
Extensive acquisition and formatting of financial variables through S&P Capital IQ Pro was 
significantly time-consuming due to limitations in the database. Hyperparameter settings for 
missForest and Random Forest were used before they were applied to the final testing data. 
Based on the results, technical variables such as price movements were most important to the 
analysts, but also variables connected to fundamental analysis. 
The conclusion was not particularly surprising. It became a new insight that the technical 
analysis would be so dominant as it was. 
The work has provided many new insights and lessons when it comes to machine learning, 
programming and data management.}},
  author       = {{Mikler, Carl-Douglas}},
  language     = {{swe}},
  note         = {{Student Paper}},
  title        = {{Hur estimeras riktkurser av analytikerna? Analys av estimerade aktiepriser på Stockholmsbörsen från år 2000 – 2024 med maskininlärningsmodellen Random Forest}},
  year         = {{2026}},
}