NHH Brage (Norges Handelshøyskole)
Not a member yet
8813 research outputs found
Sort by
Giving Eyes to Automated Valuation Models
Get PDFAbstract
Current Automated Valuation Models (AVMs) for real estate price appraisals often overlook qualitative factors, such as property condition, which can significantly impact valuation performance. This study utilizes two datasets, comprising 8,865 and 10,486 apartments respectively, to explore how incorporating image-based condition variables can enhance AVMs for apartment valuation in Oslo, Norway. By integrating images, the AVMs are effectively given "eyes," allowing them to account for the visual condition of apartments. To achieve this, a convolutional neural network (CNN) was trained on 250,000 images to classify 428,000 images into four room types: bathroom, bedroom, kitchen, and living room. Additionally, four separate CNNs were developed to evaluate the condition levels of these room types on a scale from one to five. Condition variables, including the average condition of the four rooms, were incorporated into hedonic linear regression, XGBoost, and multi-layer perceptron AVMs, which were trained both with and without these variables for comparison. Their impacts were analyzed using explainable artificial intelligence tools (XAI). The results demonstrated that including condition variables improved predictive performance across all AVMs. XGBoost outperformed the others, achieving relative improvements in MSE and MAPE of 11.5% and 4.34%, respectively, with corresponding reductions in error metrics of 0.13% and 0.45%. A near-linear relationship was observed between price and the average condition level, which had the greatest overall impact on valuation for the condition variables and was among the most important features. Among room types, bathrooms and kitchens had the strongest influence on valuation, followed by bedrooms and living rooms. These findings underscore the importance of image-based condition variables in enhancing AVM performance and offer valuable insights for researchers, appraisers, homeowners, and investors alike.Abstract
Current Automated Valuation Models (AVMs) for real estate price appraisals often overlook qualitative factors, such as property condition, which can significantly impact valuation performance. This study utilizes two datasets, comprising 8,865 and 10,486 apartments respectively, to explore how incorporating image-based condition variables can enhance AVMs for apartment valuation in Oslo, Norway. By integrating images, the AVMs are effectively given "eyes," allowing them to account for the visual condition of apartments. To achieve this, a convolutional neural network (CNN) was trained on 250,000 images to classify 428,000 images into four room types: bathroom, bedroom, kitchen, and living room. Additionally, four separate CNNs were developed to evaluate the condition levels of these room types on a scale from one to five. Condition variables, including the average condition of the four rooms, were incorporated into hedonic linear regression, XGBoost, and multi-layer perceptron AVMs, which were trained both with and without these variables for comparison. Their impacts were analyzed using explainable artificial intelligence tools (XAI). The results demonstrated that including condition variables improved predictive performance across all AVMs. XGBoost outperformed the others, achieving relative improvements in MSE and MAPE of 11.5% and 4.34%, respectively, with corresponding reductions in error metrics of 0.13% and 0.45%. A near-linear relationship was observed between price and the average condition level, which had the greatest overall impact on valuation for the condition variables and was among the most important features. Among room types, bathrooms and kitchens had the strongest influence on valuation, followed by bedrooms and living rooms. These findings underscore the importance of image-based condition variables in enhancing AVM performance and offer valuable insights for researchers, appraisers, homeowners, and investors alike
Home Sweet Home?
Get PDFTidligere nasjonale utvalg har konkludert med at Norge har en lav grad av home bias, uten å ta hensyn til påvirkningen fra Statens Pensjonsfond Utland (SPU). I denne masteroppgaven beregner vi home biasen i det norske noterte aksjemarkedet fra 1995 til 2023, og dokumenterer hvordan fondets størrelse og mandat til å investere i utlandet reduserer graden av home bias betraktelig. Våre funn viser at home bias er fire ganger så stor når man ekskluderer SPU, og utfordrer derfor antakelsen om at Norge har en lav grad av home bias. Vi skiller mellom home bias på både nasjonalt nivå, på tvers av sektorer, og i sammenligning med de andre skandinaviske landene. Vi beviser at den aggregerte nasjonale home biasen inkludert SPU i stor grad overskygger den mer nøyaktige home biasen til norske investorer. Våre funn fremhever hvordan inkluderingen av fondet i home bias beregninger kan være suboptimalt når man vurderer det norske kapitalmarkedet; Et marked fondet er begrenset fra å delta i. Vår analyse tyder på at eksisterende norsk skattepolitikk, basert på antakelsen om at home biasen er lav i Norge, kan måtte revurderes.Prior national committees have concluded that Norway has a low degree of home bias, without considering the impact of the Government Pension Fund Global (GPFG). In this thesis, we calculate home bias in the listed Norwegian stock market from 1995 to 2023, documenting how the fund’s size and mandate to strictly invest abroad considerably reduce the degree of home bias. Our findings reveal that home bias is four times higher when excluding the GPFG, thereby challenging the assumption of Norway exhibiting a low degree of home bias. We distinguish between home bias at both national levels, across sectors, and in comparison with the other Scandinavian countries. We prove that the aggregated national home bias calculation including the GPFG largely overshadow the more accurate home bias of Norwegian investors. Our findings highlight how the inclusion of the fund in home bias calculations can be suboptimal when assessing the Norwegian capital market; A market the fund is restricted from participating in. Our analysis suggests that existing Norwegian tax policy based on the assumption that home bias is low in Norway might need to be re-evaluated
The Impact of Wind Power on the Norwegian Wholesale Electricity Market
Get PDFIn this thesis, the effect of wind power production on the Norwegian wholesale electricity market was explored. To investigate the dynamics of wind power production on the wholesale electricity market, we used the pairwise price spread between the three sequential markets. We estimated a vector auto-regressive model for the five price zones in Norway to study the interrelationships between price spread, wind forecast errors, non-wind forecast errors, load (consumption) forecast errors, and net exchange. Based on the VAR model, we performed a Granger causality analysis and used Structural Impulse Response
Function (IRF)’s to study the relationships in the wholesale electricity market.
We found that wind forecast errors affect the dynamics in the Norwegian wholesale electricity market, where all of the studied variables are significantly affected. A positive (negative) shock in wind forecast error decreases (increases) prices in the intraday and regulating power markets. Interestingly, the effect is more prominent in the regulating power market than in the intraday market compared to similar studies of other Nordic countries. We argue that this can be explained by Norway’s high share of flexible hydropower production, which lowers price risk in the regulating power market, thus reducing incentives to handle imbalances in the intraday market compared to other countries.
We found two opposing effects of wind forecast error on the response of hydropower production. Increased (decreased) wind power production reduces (increases) hydropower to balance the grid, while increased (decreased) wind also increases (decreases) consumption. The size of these two effects determined the total hydropower response.
Our results support previous literature as well as discovering new dynamics of how wind power forecast errors affect the wholesale electricity market in a country like Norway with highly flexible hydropower production
Prislappen på Hans
Get PDFDenne masteroppgaven undersøker de økonomiske konsekvensene av ekstremværet Hans og de bredere utfordringene klimaendringer medfører. Formålet er å kartlegge og analysere totalkostnadene ved slike hendelser, inkludert direkte og indirekte skader, og vurdere hvordan dagens økonomiske ordninger og ansvarsfordeling kan forbedres for å møte fremtidens behov. Studien kombinerer kvalitative metoder, som intervjuer og sekundærdata, med teoretiske rammeverk som omhandler naturrisiko, endringsrisiko og økonomiske konsekvenser av naturkatastrofer. Funnene har synliggjort samfunnets sårbarhet i møte med ekstremværet Hans, med betydelige økonomiske kostnader, særlig knyttet til infrastruktur, forsikring og landbruk. Dette understreker viktigheten av forebyggende tiltak, som flomsikring og styrking av kritisk infrastruktur, som både er kostnadseffektive og nødvendige for å redusere risikoen for fremtidige skader. Analysen viser også at dagens forsikrings- og beredskapssystemer ikke er tilstrekkelig tilpasset hyppigere og mer intense værhendelser, og det er derfor behov for en gjennomgang og styrking av eksisterende ordninger. Videre fremhever studien viktigheten av samarbeid mellom offentlige og private aktører for å håndtere klimarisiko mer effektivt. Forbedret datadeling og utvikling av helhetlige modeller som integrerer både direkte og indirekte kostnader, kan styrke beslutningsgrunnlaget for klimatilpasning. Oppgaven konkluderer med at forebygging er vesentlig mer kostnadseffektivt enn reparasjon, og at Norge må prioritere klimatilpasning for å møte fremtidens utfordringer. Dette krever investeringer i forebyggende tiltak og bedre samordning mellom aktører. Videre forskning på indirekte kostnader, integrerte risikomodeller og forbedret datadeling kan bidra til å styrke samfunnets beredskap og økonomiske stabilitet overfor klimaendringer
To Growth and Beyond: An Empirical Analysis of Growth Equity Fund Returns, Characteristics, & Performance Persistence
Get PDFAlthough growth equity (GE) has emerged as the fastest-growing fund type in private equity (PE) the last decade, empirical research remains limited on how GE compares to venture capital (VC) and buyout funds. This study addresses this gap by analyzing GE fund performance relative to VC and buyout funds using a Pitchbook dataset spanning from 2004 to 2019 and drawing on established methodologies in PE research.
We find that GE funds deliver higher average returns than buyout funds but do not outperform VC funds. In addition, our results suggest that increased fund size negatively affects performance across all fund types, while increased fund manager experience has a positive effect. Moreover, we find evidence of performance persistence among GE fund managers. Overall, our findings highlight GE’s relevance as a growing fund type and provide valuable insights for investors regarding fund size and manager selection.Although growth equity (GE) has emerged as the fastest-growing fund type in private equity (PE) the last decade, empirical research remains limited on how GE compares to venture capital (VC) and buyout funds. This study addresses this gap by analyzing GE fund performance relative to VC and buyout funds using a Pitchbook dataset spanning from 2004 to 2019 and drawing on established methodologies in PE research.
We find that GE funds deliver higher average returns than buyout funds but do not outperform VC funds. In addition, our results suggest that increased fund size negatively affects performance across all fund types, while increased fund manager experience has a positive effect. Moreover, we find evidence of performance persistence among GE fund managers. Overall, our findings highlight GE’s relevance as a growing fund type and provide valuable insights for investors regarding fund size and manager selection
The Connecting Link between Organizational Incentive Theory and Sustainable Shipping Practices
Get PDFThe objective of this master thesis is to identify the main drivers and barriers to effectively incentivize shipping companies in the Norwegian shipping sector to implement sustainable and circular economy practices. Our study is limited to the short sea and the offshore shipping segment in Norway. This thesis is grounded in theoretical literature, on sustainability and Circular Economy (Altuntaş Vural et al., 2021; Okumus et al., 2023; Geissdoerfer et al., 2017), internal drivers and barriers for implementing sustainable and circular solutions (Tang & Gekara, 2020; Chang & Danao, 2017, Yuen & Lim, 2016) and external drivers and barriers for sustainable and circular solutions (Tang & Gekara, 2020; Raza, 2020; Felício et al., 2021; Skovgaard, 2014). This thesis addresses four research questions, culminated into a four-step research model designed to examine the industry's perceptions on sustainability and circular economy, as well as identify internal and external incentives, and assess how these factors are prioritized. Ultimately, these findings provide insights on how to effectively incentivize sustainability practices. To answer these research questions, a semi-structured interview method will be employed, with ten respondents from six companies within the Norwegian shipping industry. The study categorizes both internal and external incentives, revealing key factors, such as laws and regulations, customers willingness to pay, public subsidy schemes, the alignment of incentives, the influence of top management, and organizational culture considerations as key elements to implement sustainability and circular economy practices. These findings offer valuable insights into potential strategies for increasing the focus on sustainability and circular economy within the Norwegian shipping industry. However, further research is needed to substantiate and statistically validate these results. This thesis contributes to the understanding of why companies adopt sustainable solutions, identifies the incentives that influence sustainable decision-making, and explores how this knowledge can be applied to encourage greater sustainability in practice.The objective of this master thesis is to identify the main drivers and barriers to effectively incentivize shipping companies in the Norwegian shipping sector to implement sustainable and circular economy practices. Our study is limited to the short sea and the offshore shipping segment in Norway. This thesis is grounded in theoretical literature, on sustainability and Circular Economy (Altuntaş Vural et al., 2021; Okumus et al., 2023; Geissdoerfer et al., 2017), internal drivers and barriers for implementing sustainable and circular solutions (Tang & Gekara, 2020; Chang & Danao, 2017, Yuen & Lim, 2016) and external drivers and barriers for sustainable and circular solutions (Tang & Gekara, 2020; Raza, 2020; Felício et al., 2021; Skovgaard, 2014). This thesis addresses four research questions, culminated into a four-step research model designed to examine the industry's perceptions on sustainability and circular economy, as well as identify internal and external incentives, and assess how these factors are prioritized. Ultimately, these findings provide insights on how to effectively incentivize sustainability practices. To answer these research questions, a semi-structured interview method will be employed, with ten respondents from six companies within the Norwegian shipping industry. The study categorizes both internal and external incentives, revealing key factors, such as laws and regulations, customers willingness to pay, public subsidy schemes, the alignment of incentives, the influence of top management, and organizational culture considerations as key elements to implement sustainability and circular economy practices. These findings offer valuable insights into potential strategies for increasing the focus on sustainability and circular economy within the Norwegian shipping industry. However, further research is needed to substantiate and statistically validate these results. This thesis contributes to the understanding of why companies adopt sustainable solutions, identifies the incentives that influence sustainable decision-making, and explores how this knowledge can be applied to encourage greater sustainability in practice
The Global Scaling Paradox from a Market Perspective
Get PDFNorge har som mål å bli det mest digitaliserte landet i verden, og satser på en overgang fra tradisjonell eksport til digitale eksportnæringer. Likevel henger norske bedrifter etter når det gjelder å eksportere sine digitale løsninger. Dette understreker behovet for å undersøke utfordringene knyttet til internasjonal skalering for norske digitale bedrifter. En vanlig utfordring ved ekspansjon til nye markeder er å opprettholde effektiv fornyelse av bedriftens tilbud. Tippmann et al. (2023) beskriver denne balansen mellom innovasjon og ekspansjon som the global scaling paradox, og peker på strategier som multinasjonale selskaper bruker for å håndtere denne spenningen. Det finnes imidlertid begrenset forskning på hvordan små og mellomstore digitale bedrifter (SMBer) håndterer denne balansen. Denne oppgaven utforsker derfor hvordan norske digitale SMBer adresserer kravene til kontinuerlig innovasjon samtidig som de ekspanderer. Gjennom en flercasestudie undersøker vi 12 norske digitale SMBer på ulike stadier av deres internasjonaliseringsreise. Basert på våre funn, fant vi at ekspansjon og innovasjon ikke er separate krav, men heller sammenkoblede prosesser, der ekspansjon driver innovasjon for digitale SMBer. Ekspansjon krever ofte produkttilpasninger, men samtidig må SMBene opprettholde verdien av sine kjerneprodukter for å beholde eksisterende kunder og sikre stabile inntektsstrømmer. Dette førte oss til en redefinert konseptuell modell, det globale skaleringsparadokset fra et markedsperspektiv, som fremhever at kundetilbakemeldinger spiller en kritisk rolle i å opprettholde konkurransekraft i globale markeder. Denne studien gir også verdifull innsikt for både praktikere og beslutningstakere. For digitale SMBer kan systematisk håndtering av kundetilbakemeldinger forbedre produktutviklingen. Norske politikere bør også adressere finansielle og strukturelle barrierer, inkludert begrenset tilgang til kapital og restriktiv skattepolitikk, for å støtte SMBer i å oppnå bærekraftig vekst og opprettholde global konkurransekraft.Norway aims to become the most digitalized country globally, relying on transitioning from traditional export industries to digital export industries. However, Norwegian firms lag behind in exporting their digital solutions, which highlight the need to address challenges in international scaling for Norwegian digital firms. A common challenge when expanding into new markets is maintaining the efficiency renewal of a firm’s offerings. Tippmann et al. (2023) describe this balance between innovation and expansion as the global scaling paradox, highlighting strategies used by multinational enterprises (MNEs) to navigate this tension. However, limited research exists on how digital small and medium-sized enterprises (SMEs) manage this balance. This thesis therefore explores how Norwegian digital SMEs address the demands of continuous innovation while pursuing global expansion. Through a multiple-case study approach, we examine 12 Norwegian digital SMEs at various stages of their internationalization journey. Our study finds that expansion and innovation are not separate demands but rather interconnected processes, where expansion fuels innovation for digital SMEs. Expansion often requires product adaptations, but at the same time SMEs must maintain the value of their core offerings to retain existing customers and sustain their revenue streams. This dual pressure faced by SMEs led us to a redefined conceptual model, the global scaling paradox from a market perspective, highlighting the critical role of customer feedback in maintaining a competitive edge in global markets. This study also provides valuable insights for practitioners and policymakers. For digital SMEs, systematic feedback management, can improve product development. Norwegian policymakers should address financial and structural barriers, including limited access to capital and restrictive tax policies, to support SMEs in achieving sustainable growth and maintaining competitiveness globally
Investigating the Credit Spread Premium in Norway's High-Yield Bond Market Compared to the U.S. Market
Get PDFThis master thesis investigates the consistent credit spread premium in Norway's high-yield bond market compared to the U.S. market by identifying and measuring the underlying factors driving this difference. Our analysis relies on carefully filtered data mainly retrieved from Stamdata and Bloomberg and includes issue-specific details, financial metrics, and market data for bonds issued between January 01, 2011, and October 16th 2024, to ensure a good reflection of the current market dynamics.
Our research is structured around two main objectives. First, we utilize the extended Merton model by Eom et al. (2004) to calculate the credit risk component of the total credit spread for both markets. To further investigate the spread differential, we conduct regression analyses that test the impact of non-default-related variables, incorporating interaction terms to capture their unique effects across the two markets.
The main findings of the thesis are that credit risk explains approximately 54% of the total credit spread for the Norwegian high-yield bond market, while it explains 72% for the U.S. market, highlighting the greater influence of non-default factors in Norway. The extended Merton model underpredicts spreads in both markets, with an average mispricing of 225.93 bps for Norway and 75.40 bps for the U.S. Additionally, we found that liquidity is a key factor explaining the credit spread premium in Norway, contributing to 165.96 bps of the credit spread
Kjønnsforskjeller i bruk og adopsjon av generativ KI; hvilken påvirkning har opplæring, retningslinjer og ledelsesinnflytelse?
Get PDFDenne masteravhandlingen utforsker hvilke kjønnsforskjeller i bruk og adopsjon av generativ kunstig intelligens (GKI) som eksisterer blant kontoransatte i Norge i dag. Formålet med studien er å diskutere hvilken påvirkning ulike tiltak vil ha på de observerte kjønnsforskjellene. Studien baserer seg på en spørreundersøkelse med svar fra 200 kontoransatte i Norge.
Studien er utformet med bakgrunn i eksisterende forskning på kjønnsforskjeller ved bruk og adopsjon av ny teknologi, med et hovedfokus på GKI, samt hvordan tiltak som opplæring, retningslinjer og ledelsesinnflytelse kan påvirke de observerte forskjellene. Den eksisterende forskningen danner grunnlaget for fire hypoteser som vi vil konkludere med basert på diskusjon av våre funn sammen med den eksisterende forskningen på området. Sammen vil dette danne grunnlaget for konklusjonen på vårt forskningsspørsmål.
Funn fra denne studien viser at det eksisterer kjønnsforskjeller i bruk og adopsjon av GKI blant kontoransatte i Norge i dag. Kjønnsforskjellene med hensyn på bruk relaterer seg i hovedsak til at flere menn enn kvinner bruker det på fritiden, og at kvinnene i enkelte arbeidsoppgaver benytter GKI i mindre grad sammenlignet med mennene. Vi avdekker i tillegg kjønnsforskjeller med hensyn på adopsjon der kvinnene er mindre komfortable med å bruke det i arbeidslivet, mennene anser egen bruk av GKI i arbeidslivet i mindre grad som juks, og kvinnene anser egen kunnskap på området som mindre sammenlignet med kollegaene. I tillegg finner vi at kvinnene i større grad introduseres til bruken av dette i arbeidslivet gjennom andre, sammenlignet med mennene, som i størst grad introduserer seg selv for det.
I sammenheng med ønsket om å avdekke hvilke kjønnsforskjeller som eksisterer, var det viktig for oss å se på hvilke tiltak som kunne påvirke disse forskjellene. Vi fant at kvinnene i større grad enn mennene viste en interesse for opplæring og at de ville følt seg tryggere ved tilrettelegging gjennom opplæring og tydeligere retningslinjer. Til slutt fant vi at kvinnene ville blitt mer påvirket enn mennene av ledelsesinnflytelse i form av oppmuntring og fraråding, der de i større grad ville brukt GKI mer ved oppmuntring og mindre ved fraråding.
Studien har som hensikt å gi viktig innsikt til ansatte og bedrifter i Norge, i en verden hvor teknologien er i stadig rask utvikling og der GKI har blitt en stor del av denne utviklingen. Videre danner studien et grunnlag for videre forskning som ønsker å undersøke tilsvarende kjønnsforskjeller og påvirkningen av ulike tiltak
Does increased media coverage affect resident electricity demand?
Get PDFThis thesis examines the effect of media attention on Norwegian residential electricity demand. Focusing on how increased media attention might impact the price elasticity of electricity demand during the electricity crisis of 2021 – 2023. This period saw unprecedentedly high electricity prices, which became a constant focus of national news coverage in Norway. Using panel data IV-regression models with an interaction term, our thesis investigates a direct and indirect “media effect” on residential electricity demand.
Our findings suggest that although there are signs of a positive direct media effect. This indicates that rising media attention leads to higher electricity demand from Norwegian households, which is unexpected. This would indicate that as media attention on the electricity prices increased, households responded by increasing their consumption of electricity, rather than decreasing it. Simultaneously, our hypothesized interaction effect between media attention and electricity price seems to be nonexistent. Implying that increased media attention does not affect the price elasticity of electricity.
These findings highlight the complexity of how media attention influences household behavior. There might be several reasons for our surprising results, such as how media attention is measured, the availability of suitable instruments, or the impact of the Norwegian electricity subsidy.
In addition, our analysis estimates that a one percentage point increase in the price of electricity, leads to a 0.05 percent reduction in household electricity demand. Although this is higher than previous relevant studies, it does not indicate that the household electricity demand is elastic. Implying that the electricity crisis between 2021 and 2023 did not significantly alter the residential price elasticity of electricity