Paradigms in Statistical Inference for Finite Populations; Up to the 1950s

Modern sample surveys started to spread after statistician at the U.S. Bureau of the Census in the 1940s had developed a sampling design for the Current Population Survey (CPS). A significant factor was also that digital computers became available for statisticians. In the beginning of 1950s, the theory was documented in textbooks on survey sampling. This thesis is about the development of the statistical inference for sample surveys. For the first time the idea of statistical inference was enunciated by a French scientist, P. S. Laplace. In 1781, he published a plan for a partial investigation in which he determined the sample size needed to reach the desired accuracy in estimation. The plan was based on Laplace s Principle of Inverse Probability and on his derivation of the Central Limit Theorem. They were published in a memoir in 1774 which is one of the origins of statistical inference. Laplace s inference model was based on Bernoulli trials and binominal probabilities. He assumed that populations were changing constantly. It was depicted by assuming a priori distributions for parameters. Laplace s inference model dominated statistical thinking for a century. Sample selection in Laplace s investigations was purposive. In 1894 in the International Statistical Institute meeting, Norwegian Anders Kiaer presented the idea of the Representative Method to draw samples. Its idea was that the sample would be a miniature of the population. It is still prevailing. The virtues of random sampling were known but practical problems of sample selection and data collection hindered its use. Arhtur Bowley realized the potentials of Kiaer s method and in the beginning of the 20th century carried out several surveys in the UK. He also developed the theory of statistical inference for finite populations. It was based on Laplace s inference model. R. A. Fisher contributions in the 1920 s constitute a watershed in the statistical science He revolutionized the theory of statistics. In addition, he introduced a new statistical inference model which is still the prevailing paradigm. The essential idea is to draw repeatedly samples from the same population and the assumption that population parameters are constants. Fisher s theory did not include a priori probabilities. Jerzy Neyman adopted Fisher s inference model and applied it to finite populations with the difference that Neyman s inference model does not include any assumptions of the distributions of the study variables. Applying Fisher s fiducial argument he developed the theory for confidence intervals. Neyman s last contribution to survey sampling presented a theory for double sampling. This gave the central idea for statisticians at the U.S. Census Bureau to develop the complex survey design for the CPS. Important criterion was to have a method in which the costs of data collection were acceptable, and which provided approximately equal interviewer workloads, besides sufficient accuracy in estimation.

Cliffordin analyysi ja sovelluksia

Cliffordin algebrat ovat äärellisulotteisia reaali- tai kompleksikertoimisia algebroja, jotka yleistävät kvaterneja ja kompleksilukuja. Näitä algebroja on kutsuttu myös geometrisiksi algebroiksi. Tässä tutkielmassa tarkastellaan analyysiä Cliffordin algebroilla ja sen sovelluksia. Analyysi tässä tarkoittaa sitä, että tarkastellaan Cliffordin algebraarvoisia funktioita, jotka omaavat erikseen määriteltyjä sileysominaisuuksia. Sovelluskohteina ovat osittaisdifferentiaaliyhtälöt ja reuna-arvo-ongelmat. Menetelmät ovat klassisia kompleksianalyysin menetelmiä. Tutkielmassa esitellään Cliffordin algebrat yleisille neliömuodollisille avaruuksille. Keskeisiä algebrallisia ominaisuuksia ovat Frobeniuksen teoreema ja perusoperaatiot. On yleisesti tunnettua, että kvaterneilla voidaan esittää kolmiulotteisen ja neljäulotteisen avaruuden rotaatiot. Tutkielmassa esitellään, miten Cliffordin ryhmiä, jotka ovat Cliffordin algebrojen osajoukkoja, käytetään useamman ulottuvuuden rotaatioiden esityksessä. Toinen sovelluskohde on Möbius-kuvausten esittäminen Vahlenin matriiseilla. Tutkielman toisessa osiossa määritellään monogeeniset funktiot erään Diracin operaattorin nollaratkaisuina. Monogeenisten funktioiden pääominaisuus on Cauchyn integraalikaava. Välittömiä seurauksia ovat esimerkiksi potenssisarjakehitelmät, analyyttisuus, Liuvillen teoreema ja muut klassisen kompleksianalyysin tuloksien yleistykset. Toisaalta monet kompleksianalyysin tulokset eivät yleisty. Esimerkiksi monogeenisten funktioiden tulo ei ole yleisesti ottaen monogeeninen. Potenssisarjat voidaan esittää monogeenisten polynomeiden avulla. Esitämme kannan monogeenisten polynomien avaruudelle käyttäen CK-laajennusta. Cauchyn ytimen ominaisuuksien avulla tarkastelemme Diracin operaattorin reuna-arvo-ongelmia ja nk. D-ongelmaa. Käyttäen Rungen lauseen yleistystä osoitamme D-ongelman yleisen ratkaistavuuden. Toisaalta reuna-arvo-ongelman ratkaistavuus karakterisoidaan käyttäen Cauchyn ytimen reuna-arvo-ominaisuuksia ja hyppyrelaatioita. Keskeinen sovellus tuloksille on aikaharmonisen Maxwellin yhtälön reuna-arvo-ongelmien tarkastelu. Mielenkiintoista on myös, miten Diracin operaattori linearisoi Laplacen operaattorin ja aalto-operaattorin. Toisaalta Diracin operaattorin avulla voidaan ilmaista Maxwellin yhtälöt tiiviissä muodossa. Muita tuloksia tutkielmassa ovat meromorfifunktioiden määritelmä ja Mittag-Lefflerin lause. Tutkielman lopuksi tarkastellaan lyhyesti harmonisten funktioiden ja monogeenisten funktioiden suhdetta. Jokainen harmoninen funktio on jonkin monogeenisen funktion reaaliosa. Tosin monogeeninen funktio ei ole yksikäsitteisesti määrätty sen reaaliosan avulla.

Embracing Integrability in Stellar and Planetary Dynamics

Hamiltonian systems in stellar and planetary dynamics are typically near integrable. For example, Solar System planets are almost in two-body orbits, and in simulations of the Galaxy, the orbits of stars seem regular. For such systems, sophisticated numerical methods can be developed through integrable approximations. Following this theme, we discuss three distinct problems. We start by considering numerical integration techniques for planetary systems. Perturbation methods (that utilize the integrability of the two-body motion) are preferred over conventional "blind" integration schemes. We introduce perturbation methods formulated with Cartesian variables. In our numerical comparisons, these are superior to their conventional counterparts, but, by definition, lack the energy-preserving properties of symplectic integrators. However, they are exceptionally well suited for relatively short-term integrations in which moderately high positional accuracy is required. The next exercise falls into the category of stability questions in solar systems. Traditionally, the interest has been on the orbital stability of planets, which have been quantified, e.g., by Liapunov exponents. We offer a complementary aspect by considering the protective effect that massive gas giants, like Jupiter, can offer to Earth-like planets inside the habitable zone of a planetary system. Our method produces a single quantity, called the escape rate, which characterizes the system of giant planets. We obtain some interesting results by computing escape rates for the Solar System. Galaxy modelling is our third and final topic. Because of the sheer number of stars (about 10^11 in Milky Way) galaxies are often modelled as smooth potentials hosting distributions of stars. Unfortunately, only a handful of suitable potentials are integrable (harmonic oscillator, isochrone and Stäckel potential). This severely limits the possibilities of finding an integrable approximation for an observed galaxy. A solution to this problem is torus construction; a method for numerically creating a foliation of invariant phase-space tori corresponding to a given target Hamiltonian. Canonically, the invariant tori are constructed by deforming the tori of some existing integrable toy Hamiltonian. Our contribution is to demonstrate how this can be accomplished by using a Stäckel toy Hamiltonian in ellipsoidal coordinates.

Substance misuse problems during pregnancy with special emphasis on buprenorphine

Maternal drug abuse during pregnancy endangers the future health and wellbeing of the infant and growing child. On the other hand, via maternal abstinence, these problems would never occur; so the problems would be totally preventable. Buprenorphine is widely used in opioid maintenance treatment as a substitute medication. In Finland, during 2000 s buprenorphine misuse has steadily increased. In 2009 almost one third of clientele of substance treatment units were in treatment because of buprenorphine dependence. At Helsinki Women s Clinic the first child with prenatal buprenorphine exposure was born in 2001. During 1992-2001 in the three capital area maternity hospitals (Women s clinic, Maternity hospital, Jorvi hospital) 524 women were followed at special antenatal clinics due to substance abuse problems. Three control women were drawn from birth register to each case woman and matched for parity and same place and date of the index birth. According to register data mortality rate was 38-fold higher among cases than controls within 6-15 years after index birth. Especially, the risk for violent or accidental death was increased. The women with substance misuse problems had also elevated risk for viral hepatitis and psychiatric morbidity. They were more often reimbursed for psychopharmaceuticals. Disability pensions and rehabilitation allowances were more often granted to cases than controls. In total 626 children were born from these pregnancies. According to register data 38% of these children were placed in out-of-home care as part of child protection services by the age of two years, and half of them by the age of 12 years, the median follow-up time was 5.8 years. The risk for out-of-home care was associated with factors identifiable during the pre- and perinatal period. In 2002-2005 67 pregnant women with buprenorphine dependence were followed up at the Helsinki University Hospital, Department of Obstetrics and Gynecology. Their pregnancies were uneventful. The prematurity rate was similar and there were no more major anomalies compared to the national statistics. The neonates were lighter compared to the national statistics. They were also born in good condition, with no perinatal hypoxia as defined by standard clinical parameters or certain biochemical markers in the cord blood: erythropoietin, S100 and cardiac troponin-t. Almost 80% of newborns developed neonatal abstinence syndrome (NAS) and two third of them needed morphine medication for it. Maternal smoking over ten cigarettes per day aggravated and benzodiazepine use attenuated NAS. An infant s highest urinary norbuprenorphine concentration during their first 3 days of life correlated with the duration of morphine treatment. The average length of infant s hospital stay was 25 days.

Wegener's granulomatosis in Finland in 1981-2000

Objectives: Wegener s granulomatosis (WG) is a vasculitis with a predilection for the airways and kidneys. An increasing incidence and improved prognosis of WG has been shown. The aim of this study was to evaluate the incidence, clinical presentation, diagnostic delay, risk of dialysis-dependent renal insufficiency and mortality of WG in 1981-2000. Patients and methods: Data was retrieved from the Finnish hospital discharge register and hospital case reports. Patients diagnosed with WG in 1981-2000 were included, and their demographic and clinical data recorded. The patients were crossed with the national kidney dialysis register and the national mortality statistics. Results: A total of 492 patients (243 ♂ , 249 ♀) were diagnosed at a mean age of 54 years (SD 18). The incidence increased from 1.9 to 9.3/ million/ year. The median diagnostic delay decreased from 17 to 4 months. Patients presented most often with symptoms of the ear, nose and throat (ENT) (45%), lung (36%), musculoskeletal system (22%) and kidney (11%). Initial lung involvement, constitutional symptoms, high erythrocyte sedimentation rate (ESR) and high ELK scores [(number of simultaneously involved organ groups (ENT, Lung, Kidney)] were associated with a shorter diagnostic delay. Medical treatment of WG patients remained similar in the 1980s and 1990s. Almost 90% of patients received cyclophosphamide (CYC) and more than 90% glucocorticoid medication at some point during the course of the disease. Eighty-four patients (17%) needed dialysis. Initial renal involvement and elevated serum creatinine values were related to an increased risk of dialysis-dependent kidney disease. In two-thirds of the patients, renal impairment was reversible. Dialysis became chronic (>3 months) in 32 patients (6.5%). Nineteen patients (3.9%) received a kidney transplant. Altogether 203 patients (99 men, 104 women) died before 30 June 2005. WG was the underlying cause of death in 37%. The crude one-year and five-year survival rates were 83.3% and 74.2%, respectively. The standardized mortality ratio was 3.43 (95% CI = 2.98 to 3.94). Older age and elevated creatinine level at diagnosis predicted shorter survival. ENT symptoms at presentation and treatment with CYC were associated with better outcome. There was no additional risk associated with male gender or with either of the decades (1981-1990 and 1991-2000) Conclusions: In 1981-2000, the incidence of WG increased ca. 4.5-fold and diagnostic delay decreased to ca. one-fourth, reflecting increased recognition of the disease and improved diagnostic means. WG patients are at great risk of developing dialysis-dependent renal insufficiency and an increased risk of dying. During the study period the treatment of WG did not change markedly, nor did the prognosis improve.