Sökning: "Misspecification"
Visar resultat 1 - 5 av 16 uppsatser innehållade ordet Misspecification.
1. Unconventional Monetary Policy in the United States : An empirical study of the quantitative easing (QE) effects on households and firms
Kandidat-uppsats, Linnéuniversitetet/Institutionen för nationalekonomi och statistik (NS)Sammanfattning : Quantitative Easing is an unconventional instrument when conducting monetary policy with the aim of stimulating the economy. The instrument is a complementary tool when changing the nominal interest rate is no longer effective. LÄS MER
2. Analyzing the Negative Log-Likelihood Loss in Generative Modeling
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Maximum-Likelihood Estimation (MLE) is a classic model-fitting method from probability theory. However, it has been argued repeatedly that MLE is inappropriate for synthesis applications, since its priorities are at odds with important principles of human perception, and that, e.g. LÄS MER
3. Överpresterar små bolag i en sektor som strukturellt missgynnar dem? : En studie om storlekseffekten i halvledarsektorn
Kandidat-uppsats, KTH/Fastigheter och byggandeSammanfattning : Detta arbete syftar till att undersöka relationen mellan företagsstorlek och dess aktieavkastning,annars känt som storlekseffekten, inom halvledarsektorn. Vi använder oss av två portföljer bestående av de tio största och tio minsta halvledarbolagen och görutfallstestet under perioden 2004–2015. LÄS MER
4. Time Series Anomaly Detection and Uncertainty Estimation using LSTM Autoencoders
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : The goal of this thesis is to implement an anomaly detection tool using LSTM autoencoder and apply a novel method for uncertainty estimation using Bayesian NeuralNetworks (BNNs) based on a paper from Uber research group [1]. Having a reliable anomaly detection tool and accurate uncertainty estimation is critical in many fields. LÄS MER
5. Treatment-mediator interaction when estimating causal effects
Magister-uppsats, Umeå universitet/StatistikSammanfattning : In a variety of disciplines, causal mediation analysis is routinely conducted by researchers to examine the roles of variables which lie in the causal paths between the treatment and outcome variables. These effects are commonly estimated using a parametric model approach, where one fits regression models for the mediator and outcome variables. LÄS MER